Documente Academic
Documente Profesional
Documente Cultură
Right
AMC 10 Curriculum
Quinn Perian
1
Table of Contents
AMC 10 Curriculum
Introduction 3
1 Algebra 4
1.1 Statistics 5
1.2 Vieta’s Formulas 8
1.3 Sequences and Series 13
1.4 Equations in Word Problems 17
2 Combinatorics 22
2.1 Complementary Counting 23
2.2 Principle of Inclusion and Exclusion 26
2.3 Casework 30
2.4 Pascal’s Triangle 33
3 Geometry 38
3.1 Areas 39
3.2 Circles 44
3.3 Analytical Geometry 54
3.4 Three-Dimensional Geometry 60
3.5 Similar and Congruent Triangles 65
3.6 Trigonometry 73
4 Number Theory 82
4.1 Euclidean Algorithm 83
4.2 Modular Arithmetic Part I 87
4.3 Modular Arithmetic Part II 92
Solutions 96
2
Introduction
First of all, the content of this book is also available at the web address
competemath.weebly.com. At that website, you can also submit your solutions to exercises in
order to receive feedback.
This book is intended to help you prepare for math competitions. I believe that students learn
best when they are taught to understand how and why things work. I also believe that the best
way to cement a deep understanding of concepts into your mind is by working on problems. The
problems in this book are challenging and are meant to help students learn to solve problems as
well as learning the concepts. Thus, each concept taught will start with sample problems to
introduce it, as well as problems at the end for further practice. In addition, for any student using
this book, I highly recommend that you attempt all of the sample problems on your own before
reading the given solution; you'll gain a much deeper understanding of the topics, and your
problem solving skills will become much better (even if you don't solve all of the sample
problems)! Enjoy!
3
1 Algebra
1.1 Statistics 5
1.2 Vieta’s Formulas 8
1.3 Sequences and Series 13
1.4 Equations in Word Problems 17
4
1.1 Statistics
Sample Problem 1. Find the average (also called mean) of
, , , and . How can we
use this result?
We could just compute all of the sums and take the average of those numbers, but we
will leave the numbers in the forms they are above to try to see if we can find any
interesting patterns that we could generalize. Observe that
. We notice that because
(a-3)+(b-3)+(c-3)+(d-3)=0, 3 ends up being the average. Let's try to generalize this.
First of all, we note that there are terms. Thus, we can note that our average is
To make this problem easier to think about, we will write out the set of numbers in
non-decreasing order so it looks like this: _ _ 8 _ _. Next, using what we got in the last
5
problem, note that another way to say that we want to get the largest possible integer is
to say that we want the integer that is greater than 9 by the largest possible margin.
Note that 9 minus each term in our sequence adds to 0, so we want all terms but the
largest term to be as small as possible. Other than the 8 term and the largest term, we
need 1 term greater than or equal to 8 and two terms less than or equal to 8. Thus, the
minimum values of the terms other than the largest one would be 1, 1, 8, and 8. We note
that the sum of each of these integers minus 9 is -1-1-8-8=-18, so our largest possible
positive integer is 9+18 =27.
Sample Problem 4. If group A had an average test score of 90,
group B had an average test score of 80, and together they had an
average test score of 83, find the ratio of the number of people in
group A to the number of people in group B.
Using sample problem 1, we note that the total decreases from 83% equals the total
increases. Thus, we will let the scores of the people in group A be , ,
, . . ., and . Similarly, we will let the scores of the people in group B be
, , , . . ., and . Thus,
. However, using sample problem 1,
we know that because 80-83=-3. Similarly, we
know that . Thus, , so
the ratio of people in group A to group B is 3 to 7. Try to think about this solution to
generalize this strategy.
6
Exercises
1.1.1) If the median of set S is 8, the average is , it has n elements, the range is 1, and
all of its elements are integers, find n (there could be multiple possible values of n; find
them all)?
1.1.2) Whenever an element with value v is removed from a set with n elements and
average a, what does the new average become?
7
1.2 Vieta’s Formulas
Sample Problem 1. Find the sum of the roots, not necessarily
distinct, of the quadratic
When looking at this problem, the first solution that comes to mind is just to find the
roots and to add them up. By the quadratic formula, we see that the roots of the
Sample Problem 2. Find the sum of the roots, not necessarily
distinct, of the quadratic
Using the same method as we did in the previous problem, we get that the roots are
We may
wonder if this can be extended to the product of the roots.
Sample Problem 3. Find the product of the roots, not necessarily
distinct, of the quadratic
Just as before, we know that the roots of the quadratic are Next, we
multiply and use the difference of squares to find that
Now that both of these expressions
turned out nicely, we wonder if there is a nicer explanation that could generalize this to
higher degree polynomials.
8
Sample Problem 4. Relate the sum of the roots and the coefficient
of the term in (x+1)(x+2)(x+3).
To start, we will think about how to expand this in a way that may be slightly different
than the way you are used to, in order to not expand the whole thing. When we're
expanding a bunch of binomials multiplied by each other, the result will be the sum
of each possible combination of choosing one of the two terms in each of the
binomials and multiplying the chosen terms together. For example, in this situation,
we can choose either the x term or the constant term from each binomial to multiply
into a factor. Because we want the term, we need to choose the x term from two of
the binomials and the constant term from the other. We can do this in three ways,
depending on which binomial we choose a constant from. Thus, the only relevant terms
we get from expanding this are , , and . Thus, we get the
coefficient of the term to be 1+2+3. We note that this coefficient is just -1 times the
sum of the roots.
Let's look at expanding a quadratic using the bolded method to make sure you
understand it. We will expand (x+1)(x+2). We can choose either a 1 or an x to multiply
in from the first binomial and a 2 or an x from the second. Thus, we get the possible
combinations of choosing one term from each binomial to be (x)(x), (x)(2), (1)(x), and
(1)(2). Adding these together we get the expanded term to be . If you don't
understand why this method works, think about starting with the binomial (a+b). When
we multiply (a+b) by (c+d), we get (a+b)(c+d)=c(a+b)+d(a+b)=ca+cb+da+db. In other
words, we can multiply each term of our first binomial by either of the terms in the
polynomial we multiplied it by, and then we add up each case.
Sample Problem 5. Find the sum of the roots, not necessarily
distinct, of the polynomial
This time, we can't just solve for the roots and add them, so we will need to be clever.
When stuck on a problem, it is often helpful try to represent the given information in a
different way. Because we are dealing with the roots of a polynomial, we think of using
the Fundamental Theorem of Algebra. By the Fundamental Theorem of Algebra (look
this up if you want more detail, but the proof of this theorem is beyond the scope of this
9
curriculum), we have a unique factorization of our polynomial
where each is a root of .
Looking at problems 2 and 4, we think we might get
because, in both problems, we found the sum of the roots to be -1 times the second
rightmost coefficient divided by the first coefficient. We can note that the is the
coefficient of the term in the first representation of our polynomial. Thus, we want
to find the coefficient of the term in the second representation of our polynomial.
By thinking about what the polynomial would look like if we expanded it and only
paying attention to terms in the form of for some constant c, we can see that the
coefficient is equal to . This is because, of the k factors of
that can contribute an or a , exactly must contribute an x and the other
must contribute a for some in order to produce one of the terms we are looking at.
Therefore, we see that and thus,
Sample Problem 6. Generalize what we found in sample problem
4 to the other coefficients of higher degree polynomials.
By looking at our result in problem 3, we can apply similar methods by looking at the
term (think about how many non-x terms need to multiply together to form one of
those terms) for . By using our alternate method of expansion to express each
coefficient in terms of the roots of the polynomial, we get the following equations.
. . .
10
To be clear, we start with the sum of the roots, then go to the sum of every possible
product of 2 roots, then the sum of every possible product of 3 roots, and so on, until we
get to the sum of every possible product of k roots. These are formulas are called Vieta's
Formulas, and they can be used for a wide variety of problems.
11
Exercises
1.2.1) For positive integer , the roots (real and complex) of are called the
roots of unity. For example, the roots of unity are , , , and . Given that there
are distinct roots of unity (this can be proved relatively easily using the complex
number plane; you can find more extensive information on this with a quick google
search), find the sum of the roots of unity for .
1.2.2) Find the product of all possible q if a, b, and c are the roots of
and
.
12
1.3 Sequences and Series
Before we get into this lesson, you should already know the formula for the sum of an
arithmetic series and how to prove it. If not, there are many good resources online about
this formula that you can find to learn about it.
There is no clear solution at first, so we start by thinking about what aspects of this
problem make it hard. The main challenge in this problem is that we don't know the
number of terms. The other challenge in this problem is that the terms change by
multiplication, but we are trying to find the sum of the terms. In order to reduce the
number of terms to a constant number, we think about subtracting the sequence from
itself somehow. We let . Evidently, just subtracting the
sequence from itself won't help, so we think about how we could do something slightly
different. Thinking about how each term is just 3 times the previous term, we think that
maybe we should multiply the sequence by something and then subtract the product
from the sequence. It seems natural to multiply the sequence by 3 because it essentially
just makes the sequence start at and end at , while still containing the unknown
number of terms in the middle. Thus, we write .
Because this sequence is so similar to , we see that
subtracting them would get a lot of terms to cancel out. Thus, we have
.
We can generalize this example very easily by using the above process to see that
13
, going on forever, is equal to for r<1,
because, for r<1, goes to zero as n goes to infinity (this is far from a rigorous proof,
but it should give an intuitive reason because any fully rigorous proof would be far
above the level of this curriculum).
.
This problem, again, is hard because the number of terms isn't constant. Further, in this
problem, there doesn't appear to be an easy way to subtract it from itself as we did in
the first problem. However, we recall that the key to the first problem was a ton of terms
cancelling. Thus, we think of trying to rewrite this sequence to make a bunch of terms
cancel. Because each pair of consecutive terms has a number in the denominator in
common with the denominator of the second term, we think of using that. Thus, we
14
rewrite this sum as . This
method of rewriting the terms of the sequence in such a way to make almost all of the
terms cancel out and leave us with a simple to evaluate expression is called telescoping
because the series collapses like a folding telescope does. Telescoping can be done
without rewriting the terms through cleverly multiplying by a term or other methods.
15
Exercises
16
1.4 Equations in Word Problems
Rate Problems
Sample Problem 1. Car A travels at a speed of 25 mph. If car B
starts 5 miles behind car A, at what speed does car B travel if it
catches up with car A in 30 minutes?
For this problem, we can use the equation d=rt, where d is the distance, r is the rate, and
t is the time. If this equation doesn't appear to be obviously true immediately, think
about how the units will cancel out. We will want to introduce some variables to
represent what we already know and what we want to know. In general, when doing
word problems, it is often a good idea to start by defining variables. Let , , and
be the distance car A travels, the rate at which car A travels, and the amount of time car
A travels. We will define , , and similarly. Next, we will look at the information
we are given. We know that . We are also given that when ,
. Thus, we know that . By writing the equation
, we can see that . Another way to look at this
problem, that is essentially the same, but a little quicker, is to see that car B travels 5
more miles than car A in half an hour, so car B has a speed greater than car A's
speed. Thus, car B has a speed of , as we saw before. One important thing to
note, and to be careful of, is that we are using mph as speed, so is only true if all
of the units are the same, meaning we must note that we used
.
17
Sample Problem 2. Fred runs at a speed of 7 mph. Ned runs at a
speed of 6 mph. If Fred and Ned both start at the same place on a
169 mile track, how long until Fred and Ned meet for the first
time once they start running.
We will, again, use d=rt. For Fred and Ned to meet, they will have had to travel the full
distance of the track when the distances each ran is combined. In one hour, we can see
they travel a total of closer to each other. Thus, we see that 169=13t,
so they will meet each other for the first time in exactly 13 hours.
Sample Problem 3. If Fred biked up a hill at mph and biked
down the same hill at mph, find Fred's average speed across
the whole ride in terms of and .
When we first see this problem, we may think that the answer is just . However,
we can intuitively see that his is wrong because Fred is biking for longer when he bikes
at a slower rate. For this problem, we already have two variables defined for us and, as
we have seen previously, it is a good idea to assign variables to everything. Thus, we
will call the distance from the bottom to the top of the hill miles. We can then easily
see that Fred traveled miles in total. We want to find Fred's average speed, which is
imply total distance divided by total time. We have the distance, so now we just want to
find the time. For the first part of his ride, Fred traveled miles at a rate of mph.
Thus, for the first part he took hours. Similarly, for going down the hill, it took Fred
hours. Thus, Fred's total time was + = hours. Finally, we can easily
find Fred's average rate by taking the distance divided by the time to get his rate to be
18
Work Problems
Similar to our previous two problems, we will use the equation w=rt where w is the
amount of work, r is the rate, and t is the time. In this problem, we let the units for w be
houses painted, units for r be in houses per hour for one worker, and t be in hours.
Thus, , so each worker can paint of a house per hour. Thus, 3 workers can a
paint of a house in an hour. Thus, to paint 3 house, it takes 3 workers hours.
Before moving on, it is important to note that it took 7 not 3 hours slightly unintuitively.
We can notice that this task took the same amount of time as it did for 7 workers to
paint 7 houses, so we wonder if there is a nicer explanation for this. Indeed, there is.
Using the equation w=rt, and solving for each variable, it can be easily seen that when
we change the number of workers from a to b, the amount of time gets multiplied by .
Similarly, when we change the number of houses needing painting from c to d, we
multiply the time by . Thus, in the situation of the problem, we are multiplying 7
hours by . Make sure you understand what happens when you change the
number of workers and the amount of work, it can be very easily generalized to other
problems and will often pop up.
19
Ratio Problems
Sample Problem 5. In a town of 98 people, of the men and of
the women are married. Assuming each marriage consists of one
man and one woman and consists of only people in the town,
how many men are in the town.
20
Exercises
1.4.1) Albert is running a lemonade stand. He got 2 more cups than than he had ounces
of lemonade. The cooler he got could unfortunately only hold one less than half of the
number of cups he got. Finally, the cooler he had could hold one more than one fourth
of the number of ounces of lemonade he. How many ounces of lemonade did he have?
in only 3 minutes. She was then tired and decreased her speed by 1 mph for the rest of
distance to have a total time of 7 minutes. How long is the track?
21
2 Combinatorics
2.1 Complementary Counting 23
2.2 Principle of Inclusion and Exclusion 26
2.3 Casework 30
2.4 Pascal’s Triangle 33
22
2.1 Complementary Counting
Sample Problem 1. How many positive integers less than 101 are
not multiples of 3?
If we start out trying to to directly count this quantity, it quickly becomes apparent that
this is fairly hard. We find that it is hard to identify which numbers are not multiples of
3 without initially thinking about which numbers are. Thus, it would make sense to try
to think about the problem in a different way.
Sample Problem 2. How many positive integers below 100 are
multiples of 3 but not 8?
In this problem it isn't apparent which numbers are multiples of 3 but not multiples of
8. Thus, we think of using complementary counting. It is important to be careful here,
because it may at first appear that, when we count the opposite, we count the multiples
of 8 but not 3. However, we instead need to count the number of positive integers less
than 100 that are not multiples of 3 or are multiples of 3 and 8. Thus, it is important to
be careful in making sure we count all numbers not in our desired set but in our total
and to make sure that we properly determine what opposite to count. Using
complementary counting to solve the problem in this way is perfectly legitimate, but we
will show another way here for the purpose of introducing a useful concept.
23
Instead of trying to use positive integers below 100 as the total, we will instead think of
a different total. We will use the multiples of 3 below 100, because then we essentially
reduce our problem to the previous one. There are 33 multiples of 3 below 100. Next, we
need to remove the numbers of multiples of 3 that are also a multiple of 8. We can easily
see that each such number will be a multiple of 24. We can easily count that there are 4
positive multiples of 24 below 100. Thus, there are 33-4=29 positive integers meeting the
desired conditions.
24
Exercises
2.1.1) How many ways can 5 beads, 2 blue and 3 red, be arranged in a line without the
blue beads being next to each other (beads of the same color are indistinguishable)?
25
2.2 Principle of Inclusion and
Exclusion
Sample Problem 1. In a school, there are 20 student who play
sports, 15 who play musical instruments, and 5 who play a sport
and a musical instrument. How many students play a sport or a
musical instrument (or both)?
When doing this problem, we need to be careful to not over count any student. For
example, answers of 15+20=35 or 15+20+5=40 would both be wrong because they over
count the number of students doing both. If you are confused in questions like these,
especially when they get more complicated, it can often be a good idea to draw a Venn
Diagram like the one seen below.
By drawing a Venn Diagram, we can make sure to not over count any person. Once we
have drawn the Venn Diagram, as in the picture above, we can often easily find the
26
desired value. In this case, it is the total number of students, so we just add up the
numbers in the diagram to get a total of 30 students.
Drawing a Venn Diagram certainly works and is a good technique to remember when
you're confused, but it also isn't the most efficient. We would prefer if there were an
algebraic way to solve this problem without drawing out a diagram. Thus, we will try
to look at exactly what we did in drawing the Venn Diagram in order to find an
algebraic representation of our steps. In drawing this diagram, we were initially given
20 students who play sports and 15 students who play a musical instrument. We
subtracted the 5 people doing both activities from the 20 people playing sports and the
15 people playing an instrument in order to find the people doing just one activity. We
then added the 5 people back in to account for the people doing both activities. Thus,
we subtracted 5 twice and added it once, for a total of subtracting it once.
The process described above can be easily generalized to see that the total number of
elements that are in either one set or another (or both) is equal to the sum of the number
of elements in each set individually minus the number of elements that are in both sets.
Sample Problem 2. Generalize the last problem in order to find a
general formula for the total number of elements in any of n sets.
Make sure to not over count any elements. You are given the total
number of elements in the intersection of any z sets for each z less
than or equal to n.
It isn't immediately apparent where to start, so we will begin by looking at another
example. We will look at three different sets and try to find a general formula for that
case. First of all, if we just add together the number of elements in each set, we will have
counted the elements in the intersection between exactly two sets twice (once for each
set they are in). Thus, we need to subtract the number of elements in each pair of two
different sets. Next, for the elements in all 3 sets, we counted them 3 times when we
27
From the two previous examples, we may start to see a pattern emerge. It would
appear that we start by adding in the number of elements in each set, then subtract
the number of elements in each pair of sets, add the number of elements in each
triplet of sets, subtract the number of elements in each group 4 sets, and so on,
alternating adding and subtracting until we get to the elements in all n sets.
We will prove that the above method works by proving that any arbitrary element in
exactly sets will be counted exactly once, similar to how we worked out the total
number of elements in at least one set for 3 total sets. It can easily be verified that, for
any element in exactly sets , we add or subtract it times when we add or subtract
the elements in each group of sets. Thus, any element in exactly sets will be counted
While this general formula is good to know and will be useful, it is even more
important to be able to understand the derivation of the formula so that you would
know how to find the number of elements in at least of the total sets. Also,
remember that if you are ever stuck on a problem of this type, using a Venn Diagram is
always an option, even if it takes a bit longer.
28
Exercises
2.2.1) At Math Problem High School, students either play bridge, are on the math team,
play chess, or are on the science team. If students play bridge, chess, and are on the
math team; students play bridge and chess and are on the science team; students
play bridge and are on the math and science teams; students play chess and are on
the math and science teams; and students play bridge and chess and are on the math
and science teams, how many students are on at least 3 of the 4 teams?
2.2.2) How many positive integers below 101 are not multiples of 3 or 8?
29
2.3 Casework
Sample Problem 1. How many triangles are in the diagram
below?
There are no fancy techniques for this problem but to tough it out and count the
possibilities. However, this doesn't mean we have to just look for triangles, make tally
marks for each one we find, and hope that we don't over count or miss anything.
Instead, in this problem and every casework problem, the most important thing you can
do is to have an organized system through which you can count.
For this problem, there are many ways to organize your counting, none more correct
than another. However, in this solution, we will detail 1 specific method of organization.
First of all, there are 5 triangles that don't have any triangles inside of them. These
triangles are ABF, BGF, BCG, DCF, and EFD. We will proceed by casework on which
triangles contain each of these triangles. First of all, we can easily see that ABF is
30
contained in 3 triangles (excluding itself). Next, in order to not over count, we can easily
see that there are 2 triangles that contain BGF but not ABF (excluding BGF). There are 0
triangles that contain BCG but not ABF or BGF (excluding BCG). There are 0 triangles
that contain DCF but not ABF, BGF, or BCG (excluding DCF). There are 0 triangles that
contain FED but not ABF, BGF, BCG, or DCF (excluding FED). Thus, there are a total of
5+3+2=10 total triangles.
Notice how we made it easier to count the triangles, avoided over counting easily, and
we are certain that we counted each triangle, because we were organized, neat, and we
had a method. What method we use or how we organize our casework isn't that
important, as long as we are organized, neat, have a method. Also, note that other
concepts can be used in combination with casework, such as complementary counting
and the Principle of Inclusion and Exclusion, in order to reduce the number of cases or
to make the casework easier.
31
Exercises
2.3.1) What is the probability of flipping at least 2 heads if you flip 6 coins?
2.3.2) How many ways can 12 be made out of 1s, 2s, and 4s?
32
2.4 Pascal’s Triangle
This is Pascal's Triangle. Each number, other than the 1 in the top row, is the sum of the
2 numbers above it (imagine that there are 0s surrounding the triangle). We often
number the rows starting with row 0. We also often number the numbers in each row
going from left to right, with the leftmost number being the 0th number in that row. For
example, the 1st number in row 2 would be 2 (the 1 to the left of the 2 is the 0th number
in the row). Try to look for some of the many patterns in Pascal's Triangle. In this lesson,
we will prove some of these patterns.
Sample Problem 1. Prove that the kth number in the nth row is
.
33
while the others are down and to the left (if you can't see this immediately, imagine
going down and to the left n times to end up on the 0th number in the nth row). Thus,
the number of ways to get to the kth number in the nth row is because that is how
many ways we can choose which k of the n moves down are also to the right.
It isn't immediately apparent what the coefficients could be or how to find them, so we
will start by looking at some examples.
34
Does this look familiar? It appears that the coefficients are exactly the
numbers in the nth row of Pascal's Triangle.
We now want to figure out how to prove this. As we just saw in the last problem, we
can rewrite the numbers in Pascal's Triangle as binomial coefficients. We will thus
rewrite the coefficients of our expanded polynomials to see if we notice anything.
To start, we will note that the coefficient of the term is the same as the coefficient of
the term. It would also appear that the coefficient of the term in is
Thus, we conjecture that the coefficient of the term in is
There is a fairly intuitive explanation for why this is true. When we expand
we are essentially adding together all the combinations of choosing an x or a y from
each of n terms being multiplied together. Thus, we are essentially just adding together
35
each of the ways to choose which k of the n terms contribute an x. This also
provides a nice explanation for why the and terms have the same coefficient:
using what we just discovered, the coefficients are and respectively (the
power of the x and y term must clearly always add up to n using the explanation we
gave earlier of how to expanding ) and = (if this last statement isn't
clear think about how choosing k things to include is equivalent to choosing n-k things
not to include). This statement about the coefficients of a binomial raised to a positive
integral power is called the Binomial Theorem.
Note: The Binomial Theorem does in fact work for a binomial raised to any power, but
the proof is much too complicated to cover in this curriculum, and it far beyond this
level of math. This extended binomial theorem also uses a definition of binomial
coefficients that is adapted to be used with non-integers and negatives. If you want to
find out more, a quick google search for Extended Binomial Theorem or Generalized
Binomial Theorem should do. This theorem can also be generalized beyond binomials,
where it is called the Multinomial Theorem. A quick google search for Multinomial
Theorem should give information about that.
36
Exercises
2.4.1) Prove that the sum of the numbers in the nth row of Pascal's Triangle is .
37
3 Geometry
3.1 Areas 39
3.2 Circles 44
3.3 Analytical Geometry 54
3.4 Three-Dimensional Geometry 60
3.5 Similar and Congruent Triangles 65
3.6 Trigonometry 73
38
3.1 Areas
Sample Problem 1. Find the area of a triangle with perimeter p
whose inscribed circle has a radius of r.
Just like most geometry problems, we will start by drawing a diagram.
39
From this diagram, it should be clear that the lines from the incenter to the vertices of
the triangle split our original triangle up into 3 smaller ones. Using the formula Area=
, we get the area of our smaller triangles to be , , and . Thus, our original
The formula for the area of a triangle given above is just one of the many helpful
formulas. Below, we will provide a list of formulas to know and a brief idea of how to
prove each. We will let [ABC] be the area of the triangle, r be its inradius, s be its
semiperimeter, R be its circumradius, and a, b, and c be the lengths of its sides. We will
also let the triangle have angles with measures of A, B, and C where angle A is opposite
side BC, angle B is opposite side AC, and angle C is opposite side AB.
40
Formulas:
● [ABC]= (draw a rectangle around the triangle and look at the two
parts the altitude splits the triangle into separately)
● [ABC]= (proved easily using the Extended Law of Sines, motivation
being that we see the circumradius in the formula)
Sample Problem 2. Find the length of the altitude to the longest
side of a triangle whose side lengths are are 4, 13, and 15.
Because we know that Area= , we could solve for the length of the altitude to the
longest side if we knew the area of the triangle (b=15, h is the length of the altitude we
want to solve for). However, using Heron's Formula, we can find the area of the triangle
41
This problem introduced the important concept of finding the area of a triangle as an
intermediate step, then using the area and other knowledge to find another value. This
is where knowing all of the formulas can be helpful because you will often need to use
two formulas -- one to find the area and one to use the area you found -- and you need
to know the right formulas for the information you are given.
42
Exercises
3.1.1) Find the ratio of the inradius of a triangle to the inradius of the triangle formed by
connecting that triangle's midpoints.
43
3.2 Circles
We will start this lesson with a list of important theorems and what they say. Unlike in
most lessons, we will not fully prove all of them here, as there are a great many of them,
so we encourage you to try to prove the remaining theorems yourself and to look up
proofs online if you can't. In addition, because there are so many theorems and we want
this page to be organized, the lesson will be presented slightly differently. Even though
all of these theorems aren't in sample problem format, we encourage you to attempt to
prove them yourself before reading the solution, and we encourage you to try to look at
the methods used to prove these theorems as general strategies that can be used in
many problems involving circles (ex., extending radii out to tangents is always a good
idea). Finally, the names of these theorems are not that important to memorize, so long
as you are familiar with what the theorem does.
44
Inscribed Angle Theorem
We can prove this theorem by drawing the diameter of the circle through the vertex of
the inscribed angle and doing casework on if the endpoints of the arc are on opposite
sides of the diameter, one is on the diameter and the other isn't, or if both are on the
same side of the diameter. The first to proofs are fairly trivial and for the third proof we
45
can use our second proof to make it fairly easy as well. We encourage you to work out
the details yourself.
Some theorem states that the measure of the angle a chord forms by intersecting a
tangent at the point of tangency is half the angle of the arc in intercepts.In the diagram
to the left, this would mean that .
To prove this theorem, we will start by noting that CB is clearly the diameter of the
circle because it is perpendicular to the tangent. Thus, the measure of arc BE is equal to
. Note that, because triangle BAE is isosceles, .
46
Thus, . However, this means that .
Thus, because BE was an arbitrary chord, we have proved the desired theorem.
Another theorem states that the measure of the angle that two secants form where they
intersect is equal to half the difference in angle measures of the two arcs in the circle that
the secants intersect. For example, in the diagram to the left,
.
We will prove this theorem with the configuration where the center of the circle is in
between the secants and other configurations can be handled similarly (although we
encourage you to prove them on your own). First of all, it is apparent that quadrilateral
FAED has two angles that we want to know about, so we want to find all of the angles
in the quadrilateral in terms of angles we want to end up using. Looking at angles AFD
and DEA we see that we can easily relate them to angles GAE and HAF using the fact
that triangles GAE and HAF are isosceles. Next, we note that even though GAE and
HAF aren't angles we want to end up in our answer, the fact that
47
could help us get our answer
in terms of the angles we want. Note that
. Similarly, .
Thus,
.
The same theorem holds true for the a tangent and a secant that intersect outside of a
circle as well as a two tangents that intersect outside of a circle. The theorems say
essentially the exact same thing and can be proved in a very similar way.
A theorem states that when two chords intersect each other inside of a circle, the
measure of the angle of their intersection is equal to the average of the measure of the
arc that the angles intersects. In the diagram to the left, this would mean that
.
48
You will prove this theorem in the exercises.
Power of a Point
This theorem has a different definition than we will cover here, but as far as your use of
it will go, this definition will serve to be all you need. The theorem has several parts.
In this case, the theorem states that if we draw any chord through a point in a circle, The
product of the distance from that point to the two intersections of that chord with the
circle will stay constant. In the diagram, the theorem tells us that
.
To prove this theorem, we can just draw two arbitrary chords through an arbitrary point
and prove that these for these to chord the theorem holds. This will prove the whole
theorem because this means that for any point, if we have one chord, the desired
49
product will be equal to the product for that chord in any other chord that we draw. To
prove that this holds, we will use the picture to the left as our diagram. We can see that
triangles EBF and CDF are similar using the fact that angle BEF and DCF both intercept
the same arc-and are thus congruent by our previous our previous theorem and angles
EBF and CDF are similarly congruent. Thus, we get
.
50
51
To prove this case, we will introduce a concept called the power of a point, the very
concepts this theorem is named after. The power of point P with respect to circle O is the
distance from P to O squared minus the length of the radius of circle O squared. Note
that this quantity can be negative when P is inside circle O. For this proof, we will show
that both of our desired expressions are equal to the power of point D with respect to
circle A. First of all, by the Pythagorean Theorem, we have
. Thus, is equal to the power of
point D with respect to circle A. Next, note that
. Thus, is equal to the power
of point D with respect to circle A and we are done. Note that this implies that if two
tangents intersect at a point outside of a circle, then the distance from that point to the
points of tangency is the same for each tangent.
52
Exercises
3.2.1) Prove Theorem 3 (its diagram is labeled Thm. 3 in red).
53
3.3 Analytical Geometry
Before we start the lesson, we will go over when to use analytic geometry and when not
to. Analytic geometry rarely produces pretty solutions and is prone to arithmetic errors
and the like in the algebra. Thus, we encourage you to not jump to analytic geometry to
solve ever geometry problem. Now, this isn't to say that you should never use analytic
geometry. It is a very useful tool to have and can solve many problems easily that may
have otherwise been very hard. Analytical geometry is especially effective when there
are lots of lengths given, several right angles, and the intersection of lines. However,
even on most problems that do seem to fit these conditions, we encourage to try to find
a non-analytic (synthetic) solution first, as these solutions are often much quicker.
We will start by introducing a theorem which we will not prove. There is a proof using
induction and the cross product which is too advanced for the curriculum. However,
this theorem is quite necessary for many problems that use "coordinate bashing" as their
solution.
54
The Shoelace Formula
This formula provides a way to find the area of any polygon given its vertices in the
order they are connected. The best way to explain it is to show a visual.
55
To make this clearer, we will go over an example. Say that we had a triangle with
vertices (1,-2), (7,8), and (1,1). Note that if this weren't a triangle we would have to draw
a quick sketch to get the order of the points, but for a triangle any ordering will be
correct because each point is connected to the other two. Thus, we write out the points
in order.
(1,-2)
(7,8)
(1,1)
(1,-2)
Sample Problem 1. Find the area of triangle ABF given that E is
56
In order to show how to coordinate bash a problem, the best way is to show an
example. This problem's solution will contain many important techniques that you
should remember.
First of all, we will start by setting our coordinate axis. We let A=(0,0), B=(1,0), and
D=(1,0). Thus, E=(1,0.5). Next, using the shoelace formula, we know we can find the
desired area if only we had the coordinates of F. However, we know that we can do this
by finding the intersection of two lines. We can note that line AE has the equation
. We can also note that line BD has the equation . Thus, setting the
two lines equal and solving we get that their intersection is . Now, we could use
the shoelace formula to find the triangle's area, but in general we should look if there is
a quicker way first. In this case, now that we have the y coordinate of F, we know that
AB has length 1 and the height draw to AB has length 1/3. Thus, triangle ABF has area
. An important note is that while coordinate bashing, we can use other
formulas or some synthetic (non-coordinate) techniques to reduce our calculations.
As you saw in this problem, the key to analytical geometry is to find the coordinates of
points. you saw several useful methods of how to do this in the problem, but we will
create a more extensive list below for you to use when solving problems.
● Find the equations of two lines (or any other shape) that the point is on
and compute the intersection of the lines to find the point.
● Find a point that the point you are given is directly above or directly
horizontal to for which you know the distance between these two points.
Further, if you know that a point is some number to the left or right and
some number up or down from another point whose coordinates you
know, you can easily find that points coordinates.
● If a point is the midpoint of a line segment, then its coordinates are the
average of those of that segment's endpoints.
57
There are many other ways to find a point's coordinates, but the three listed above are
the ones that come up the most often.
58
Exercises
3.3.2) Find the Area of triangle ABE given that E is the intersection of the quarter-circle
centered at B through A and line BD.
59
3.4 Three-Dimensional Geometry
Sample Problem 1. Find the volume of the octahedron formed by
connecting the centers of the sides of a cube with edge length 1.
At first it seems we have no idea where to start. We don't even know how to find the
volume of the if we find one of its side lengths (we need two). Thus, we want to try to
think about what information we do have, preferably in 2 dimensions to make it easier
to understand. The base (the middle square-shapes part) of the octahedron seems
important, so we will look at a cross-section of the plain that it is on. When we do that,
will will see that it is simply a square with a smaller square inside it that is formed by
connecting the midpoints of the sides of the square (the smaller square is the base of the
octahedron). The larger square clearly has side lengths one (as the cross-section was
parallel to a face of the cube), so we find that the base of the octahedron has an area of
1/2. Next, we can split the octahedron into two pyramids with the same base and each
60
having a height of 1/2. Using the area we just found of the base, we get each pyramid to
have a volume of , so the total volume desired is .
● Cube: V=
● Sphere: V=
Other than these main volume formulas, most other shapes can be made of these
shapes, so you should be able to derive the volume without a formula.
61
Sample Problem 2. Find the volume that the shape below covers
when rotated around AB.
When we imaging the shape that this forms it looks like two circles, one with radius 3
and one with radius 1, that are 1 unit apart and how their circumferences connected. We
don't immediately know how to find the volume of the shape, but it should remind you
of a cone. In addition, we know that when you rotate a right triangle around similar to
how we rotated the above shape we get a right cone. However, comparing the shape
above to a triangle gives us the idea that for some point E on line AB we could draw EA
and EC to form right triangles EAC and EBD. In other words, the shape above is just
triangle EBD minus triangle EAC for some point E on line AB. However, this would
also mean the the 3d shape created by rotating the above shape is just the shape created
by rotating triangle EBD minus the shape formed by rotating triangle EAC. However,
the shape formed by rotating a triangle is just a cone, and we know how to find the
volume of cones! Thus, we just need to find the the distance EA to get the height of both
cones. Notice that for E, C, and D to be concentric we would need triangle EAC similar
62
The technique used here of thinking of one shape as a larger shape minus a smaller one
is important to remember and can be used to express many complicated 3d shapes in
terms of simpler ones we are more familiar with.
63
Exercises
64
3.5 Similar and Congruent
Triangles
Sample Problem 1. Given that AB = AC and that BD = DC, prove
that AD is perpendicular to BC.
65
essentially the same and can be one can me moved onto the other with only
translations, rotations, and reflections and all corresponding sides and angles are equal.
Note that when stating a congruence, the order of the points in each shape matter
because they are used to show how the two shapes map to each other (the first points
named by the first letter in the name of each shape map to each other, and so on). Thus,
we see that as desired, and we are done.
The claim that these two triangles are congruent may not seem obvious at first. Let's go
into more detail to see how we could prove that. Here, we will prove it by showing that
given three side lengths, any triangle constructed with those side lengths must be
congruent. To do this we will start with translate and rotate any triangle satisfying these
conditions such that the side with the longest length is mapped onto AB. From here, we
will look at all the possible triangles that could have been created. If a triangle does
have the above three lengths and longest sides at AB, then we have two possibilities in
deciding the length of the side that comes out of A. From there, we draw a circle from A
with a radius equal to the length of the side coming out of A and do the same for point
B. We thus see that the third point of our triangle is the point on the intersection
between these two circles, leaving two possibilities for where the point is. However,
because we could switch the lengths of the sides coming out of A and B, we see that
there are actually 4 possible locations for the third point of our triangle.
66
Finally, as seen in the diagram above, once we have one of the triangles corresponding
to a possible location of the third point, we can reflect it over AB, the perpendicular
bisector of AB, and both lines, in order to produce three more, for a total of 4, congruent
triangles corresponding to a possible location of the third point. However, because there
are only 4 possible locations for the third point, as described above, we have just
covered each of these possible locations and showed that each creates a triangle that can
be mapped on to each other triangle with only reflections. Thus, we have proved that
any triangle with three side lengths is congruent, and we are done. If you don't
understand this proof, go over it again until you do, because it is very important.
Now that you've seen that we can prove two triangles congruent without explicitly
proving all of their sides and all of their angles congruent, you may wonder if there are
other sets of givens that can prove two shapes congruent. Indeed, two triangles can be
proved congruent in several ways, which will be listed below. For each such method of
proving two shapes congruent, try to convince yourself that any two triangles sharing
these characteristics must indeed be congruent. Finally, try to remember the list given
67
below. Congruent triangles are the key to many geometry problems, so you will need to
know what to look for to prove two triangles congruent.
● Two triangles are congruent if they have a right angle and their
hypotenuses as well as one pair of corresponding legs are congruent
(called HL congruence).
● Two triangles are congruent if two angles and the side between them are
congruent (called ASA congruence).
68
Sample Problem 2. Using the given lengths in the diagram below,
find AD.
For this problem it isn't immediately obvious what to do, so we will start by seeing if we
can find any additional lengths using the given information. Because ABA' is a right
triangles, we can use the Pythagorean theorem to see that AA' = 5. Seeing that we now
have a lot of information about lengths for triangles ABA', we will try to see if we can
relate this triangle to some triangle that contains AD in order to try to find the length of
AD. Notice that because angle A is congruent to itself, , and
the angles of a triangle add to 180 degrees, the angles of triangle ACD are all congruent
to corresponding angles of triangle ABA'. Thus, these two triangles seem highly related,
but they definitely aren't congruent. These triangles are similar triangles, which means
that you can move one on to the other using translations, rotations, reflections, and
scaling/dilation. We can see that, by changing the length of each side of triangle ACD
by moving points C and D along sides AB and AA' respectively, we can essentially map
triangle ACD onto triangle ABA' because CD is parallel to BA'. This transformation is
called a dilation. Because CD maps to BA', we can see that this dilation mapping ACD
to ABA' doubles the lengths of each side of triangle ACD. You can see that AC and AD
must be scaled the same amount because CD is parallel to A'B, so, using the
Pythagorean theorem, we see that doubling A'B requires doubling the other lengths. In
69
order to generalize this for other problems that aren't just right triangles being dilated,
think of splitting triangles into two right triangles, or expressing an acute triangle as one
right triangle minus another, as seen in the diagram below. Thus, we can conclude AD
is half of AA' and has a length of 2.5. Make sure you understand this solution, because it
is very important.
There are several ways to prove two triangles similar. For similar triangles ABC and
DEF, it can be shown that AB/DE=BC/EF=AC/DF and that angles A and D, B and E,
and C and F are all congruent. Knowing that two triangles are similar can be used to
find lengths or angles, as in the problem above. Thus, it is important to know the
criteria for two triangles to be similar. A list of ways to show two triangles similar is
given below, and for each way you should try to convince yourself that the given
conditions do indeed make two triangles similar.
● Two triangles are similar if all three of their corresponding pairs of angles
are congruent (two pairs of corresponding angles being congruent is
enough because all angles of a triangle add to 180 degrees). This is called
AA similarity.
● If three pairs of corresponding sides all have the same ratio between the
two triangles, then the triangles are similar (called SSS similarity).
● If two pairs of corresponding sides have the same ratio between triangles
and the angle between them is the same for each triangle, then the two
triangles are similar (called SAS similarity).
70
Note that the list of ways to show two triangles similar is very similar to the ways to
show two triangles congruent, suggesting a similarity in proofs.
71
Exercises
3.5.1) Prove that the ratio between each pair of corresponding sides is the same for two
triangles who have all of their corresponding angles congruent.
72
3.6 Trigonometry
We will start this lesson off with some definitions of basic trigonometric functions.
The Sine
The sine of an angle is written as and is the distance from the intersection of
the radius of the unit circle that has been rotated counter clockwise about the origin
from the x-axis and the diameter of the unit circle to the x-axis (as can be seen in the
diagram above). Note that when this altitude to the x-axis is below the x-axis the sine of
the angle is negative. When is between and or and (if you don't
know what the radian is you should look it up because it is used very commonly), then
is positive. In addition, when is between and , can be viewed in
the context of a right triangle as the ratio of the length side opposite the angle to the
length of the hypotenuse (think about how the radius of the unit circle is the
hypotenuse of the triangle in the first definition and how from there we can scale it up
for larger hypotenuses without changing the value of the sine).
The Cosine
The Tangent
The tangent of an angle is written as and is the length of the line segment
perpendicular to the radius of the unit circle that has been rotated counter-clockwise
about the origin from the x-axis with endpoints on the x-axis and the intersection of the
73
aforementioned radius and the unit circle's diameter. The tangent is negative when
exactly one of the sine cosine is negative. The tangent can also be seen as . Thinking
about the right triangle definitions of sine and cosine, we can get that for angles
between and , the tangent in a right triangle is equal to the ratio of the side
opposite the angle to the side adjacent to the angle.
There are a few important trigonometric identities and values that you need to know.
This list will not include the sum and product identities because they are not needed on
the AMC 10, but you should look them up if you are interested: they provide proofs for
some trigonometric ratios that you should memorize. Also note that the square of a
trigonometric value is placed before the parentheses. For example, we write and
not .
●
●
●
●
●
●
●
74
●
●
●
●
●
●
For this problem, when we seen that one angle is minus another, we think of the
angles of a right triangle, especially because right triangles have so much to do with
trigonometry. By looking at a right triangle and seeing how two two non-right angles
add to , we can easily confirm this identity to be true for (
). We will proceed by casework on the quadrant that angle is
in. If it is in the second quadrant, then we can say for .
Thus,
.
For the third and fourth quadrants, note that
.
Thus, by building on each previous result, we are done.
75
Now, you may be thinking that it this is nice and all and will help find lengths in right
triangles, but how often do we really use right triangles? Well, we do use right triangles
pretty often, but we can extend this beyond right triangles. We can use trigonometry to
help us in any triangle.
Sample Problem 2. In triangle ABC let BC=a, AC=b, AB=C,
, , , and be the
radius of the circumscribed circle. Prove that
.
Knowing that we are dealing with right triangles and that we are dealing with a
circumradius, we will want to draw the circumcircle of our triangle (remember that
because any angle is half the arc it intersects, any angle that intersects the endpoints of a
diameter will be right). There doesn't appear to be any reason to to draw the diameter
through any vertex of the triangle as opposed to another one, so we try drawing the
diameter through the middle vertex. We connect the point the point on the opposite
76
endpoint of the diameter we just drew to one of the vertices because we then get an
angle intersecting the same arc as one angle of our triangle does. From here we can see
that . We can also see that because the angle
intersects the diameter of our circle. Thus, we can get
.
For any acute triangle, we would simply repeat this process for each angle to get the
desired result. However, because we have an obtuse triangle, we need to do something
different for the obtuse angle, . What we did worked well, so we want to do
something similar. We want to have an angle that is at least intersecting the arc opposite
the arc intersects (using the fact that an inscribed angle is half the measure of the
arc intersects we get that the angles add to , and we know that
). Similar to how we did last time, we connect the point on the
opposite endpoint of the endpoint we just drew to point . We again have that angle
77
78
Another relationship we can get using trigonometry is that, using the same labeling
system as the previous problem, . This is called the Law of
Cosines. We will not prove this theorem, but we encourage you to try to prove this on
your own (Hint: What theorem does this look like? What do we always often use in
trigonometry?) or to look up a proof.
79
Exercises
80
81
4 Number Theory
4.1 Euclidean Algorithm 83
4.2 Modular Arithmetic Part I 87
4.3 Modular Arithmetic Part II 92
82
4.1 Euclidean Algorithm
Sample Problem 1. What is the greatest common divisor of 388899
and 388877 (find it without a calculator)?
We want just go ahead and factor both numbers and compare factorization, but clearly
that won't work. Who would want to factor those huge numbers? There must be a
better way. Well, these two numbers do appear to be fairly close together, so maybe we
would do something with their difference, or some other smaller quantity. Well, if
, then we get that and for
integers , and .
But this gives us exactly what we wanted! This lets us realize the
because . Thus,
we have
. Now,
we simply can check to see if 388877 is divisible by 2 or 11. It is clearly not even, and
using the trick to see if a number is a multiple of 11 we find that 3+8+7-8-8-7=-5, so
388877 is not a multiple of 11. Thus, these two numbers are relatively prime or, in other
words, have a greatest common divisor of 1.
This method that we just saw seems pretty useful in helping us find the greatest
common divisor of two large numbers. But, what if our numbers don't have as small or
easy to test out of a difference? Let's try, for example, finding the greatest common
divisor of 71 and 193. Now, these numbers are smaller and we probably could factor
them, but the point here is to explore our previous method and to see if we can take it
all the way to the answer that we want. We will start the same as we did last time.
. However, the first number is still bigger than the second
number. This brings us to an important concept: using the exact same reasoning as for
one subtraction, we don't just have to subtract once; we can subtract one number from
the other until our first number is smaller than the second! In other words, we can
replace a number with its remainder when divided by the other number (note that this
83
just repeated subtraction, and sometimes repeated subtraction is easier than actually
trying to divide the number). Therefore, we have .
.
Now, we could stopped this process much earlier and easily seen our answer, but we
continued it on until we had a 0 to show something: we did end up being able to get a 0.
In fact, if , then we can get use our method to get to
(remember that 0 is a multiple of every number because any number times 0 equals 0).
Now, you might be thinking, that's nice and all, but why do I need to know that we can
always reach a zero to find the greatest common divisor of two numbers? Couldn't I
have stopped much earlier, have done less work, and have gotten answer just as easily?
The answers to these questions are that you don't need to know this to find the greatest
common divisor, yes you could have stopped earlier, yes you could have could have
done less work, and yes, if you wanted to find the greatest common divisor, you
probably should have stopped earlier. But, isn't finding the greatest common divisor
exactly what this algorithm was designed to do? The answer, again, is yes, that is what
we designed it to do, but just because we designed an algorithm to do one thing doesn't
mean it can't do another (now this isn't to say that you shouldn't use this algorithm to
find the greatest common divisor of large numbers; you should, as this algorithm makes
it much easier). We will see another way to use this algorithm we just discovered, which
is called the Euclidean Algorithm as you may have guessed. in the Diophantine
equations lesson.
Just to make sure you understand the Euclidean Algorithm fully, we will go through
another full example of it.
84
85
Exercises
4.2.1) Using the Euclidean Algorithm, express 1 in the form for integral
(not necessarily positive) and (do not guess and check; use the Euclidean
Algorithm in a way that you are guaranteed to find a solution).
86
4.2 Modular Arithmetic Part I
We will start with definitions. First of all, and most basically, we will go over the
definition of congruence in a modulo. We define to mean that is
divisible by . We say this as is congruent or equivalent to in . In other
words, it means that and have the same remainder when divided by . For
example, because 10, 4, and 1 all have a remainder of 1 when
divided by 3. We will talk about multiplication, division, addition, subtraction, and
exponentiation in the lesson. Note that we let as well. We call
the modulus.
Sample Problem 1. If some number of people are separated into
rows of 10, there are 3 left over. How many people are left over if
3 more people are arranged into the rows of 10? What does this
have to do with the definition we just discussed?
In this problem, it is clear that we aren't doing anything to the rows of 10, but just
adding 3 more people to the people left over. Thus, we have 6 people left over. This
seems to be common sense, and it is. Let's see how this applies to modular arithmetic.
First, we will start by rewriting the problem in terms of the terminology we just saw. We
will let the number of people be . Thus, we see that . Rewriting what
we found out about adding in 3 people, we see that .
Now, this is no proof that addition works as normal in modular arithmetic, but it should
be fairly clear from thinking about how we solved this problem that you can add
anything to both sides of an equivalence.
87
Sample Problem 2. Does multiplication work like it normally does
in modular arithmetic? In other words, can you multiply both
sides of a congruence by the same thing in modular arithmetic?
How about exponentiation?
We can fairly easily see that multiplication does indeed work as normal in modular
arithmetic. Too prove this we just need to remember that multiplication is just repeated
addition. Because we just saw that addition works in the previous problem. Using
similar reasoning, we can see that exponentiation works because it is just repeated
multiplication.
You might think that the answer is yes, but having fractions sometimes result from
divisions should give you a pause. After all, the very definition of modular arithmetic is
based on working with solely integers. The answer to the question ends up being, yes,
kind of, and weirdly. Let’s look at a few examples to see how we think division would
work. Let's say that we have . If we divide both sides of the
congruence by 7, we get . While may appear to work
at first, and it is a solution, there are other solutions that work, such as
. Thus, unlike the other operations we saw, does not imply
(at least not always) even though does imply
. Also, be careful in that dividing doesn't always produce a true
statement. For example, , but . In addition, division
seems weird without being able to do it to every number (we can't get fractions), so we
can conclude that division, in the traditional sense, does not work in modular
arithmetic.
We did see that one of the divisions produced a true statement, and if we want to solve
equations similar , we will need to be able to "divide." Let’s try to
think about what division does in order to try to come up with something that acts
similarly. First of all, let's look at means in algebra. using our earlier
definitions. Thinking about how it means that both sides of the congruence have the
same remainder, we can rewrite our equation as 7a=7+49y (y is integral, but not
88
necessarily positive). Thus, when we divide this equation by 7, we get a=1+7y. In other
words, by converting back to modular notation, we get . Generalizing
what we just saw, we get one form of "division" in modular arithmetic:
implies if is a divisor of , , (otherwise we get fractions and
fractions don't work with remainders and modular arithmetic).
Now that we have defined "division" in some cases, what will we do if all of the
numbers don't share a common divisor? For example, how do we find all of the
solutions to or .
We will look at cases similar to the former example first; we will look at cases of
where and have a common divisor that does not have. If this is
the case, then we can rewrite the congruence as where , ,
and are constants and is not a divisor of . We will rewrite this with plain algebra as
as we did before to get
. However, this
is a contradiction because we defined and such that is not a divisor of . Thus, if
the coefficient of the variable and the modulo share a common divisor that the other
number in our congruence doesn't, the congruence has no solution.
89
expanded upon in the Diophantine Equations lesson. However, for the AMC 10, the
best method of finding inverses is usually just to guess.
Note that taking an nth root is similar to division in that there are often multiple
solutions, however there is no nice way to find the nth roots other than guess and check.
90
Exercises
4.2.1) Prove that there exists an inverse of in where by
showing an algorithm to find the inverse (Hint: what does gcd(m,n)=1 remind you of?
How can you rephrase the inverse of a number using a purely algebraic form to relate
it?).
91
4.3 Modular Arithmetic Part II
So far, we have introduced the basics of modular arithmetic as far as simple operations
and solving 1 variable linear equations goes. We will now look at how modular
arithmetic can be used in harder problems which have a less obvious use or solution.
Sample Problem 1. If a class is arranged into rows of 10 people,
there are 3 people left over. If the same class is arranged into rows
of 13 people, there are 8 people left over. What is the minimum
number of people that could be in the class?
For this problem, you could solve it fairly easily by guessing and checking. However,
we will instead find a way to use modular arithmetic to systematically find a solution so
that when working with more constraints (more than just the 2 given) and guessing
becomes harder, you will have a way to solve the problem. In addition, it would be nice
if we could find all of the solutions. To start, we will let the number of people in the
class be . Thus, writing the information we have using modular arithmetic, we have
and . It doesn't immediately appear obvious what to
do here, so let's think about what ideas we may have. Let's think about what we would
do if we wanted to guess and check to find the solution to this problem. We would
probably start with 3 and check if it satisfies the condition (or start with 8 and
check the condition, it doesn‘t really matter). Because this doesn't work, we
would repeatedly add 10 to the 3 to keep the condition satisfied and
continually check if the number we get satisfies the condition until we found
a number that did. In other words, we would have tested numbers in the form of
to see if they are congruent to . Well, we already know how to
solve linear congruences, so why don't we just plug in the restriction into
our ? We can, and it will get us exactly to where we wanted doing
pretty much the same thing as we did when guessing and checking just without the
guessing and checking. To finish of this problem, we plug our new expression in for a to
get . We then subtract 3 from both sides to get
, and because 10 and 13 are relatively prime we can find the inverse
of 10 in by guessing if the numbers are small or using the Euclidean
92
Algorithm (usually better for big numbers) to get
. Now, be careful not to just assume the
smallest value we could get is 7, because that would be wrong. Remember that we are
trying to minimize not . From here, we can write from our modular
equation, and plugging this into , we get , or writing this
in modular form, we get that our solutions are integers that satisfy
(note that makes sense because it doesn't change the value of a number in
or to add 130). Thus, our answer is 73.
Sample Problem 2. How would we solve a system of three linear
congruences using the process we just made? How about a
system of more than three linear congruences? How does the
Chinese Remainder Theorem work with more than 2
congruences?
Thinking about how we solve a system of more than two linear equations, we can
conclude that we can just solve the congruences by taking 2 at a time. In other words,
we take any two congruences, solve them to find one congruence that represents the
solutions. We then take the new congruences and a another congruence, solve those two
congruences, and so on until we have only one congruence left that satisfies all of our
original ones. Thus, we can fairly easily generalize the Chinese remainder therefrom by
applying it to each system of two congruences that we work through to see that we will
get a solution if the moduli of all of the congruences are relatively prime, and in that
case the modulus of the solution will be the product of all of the moduli of our
individual congruences. Now, solving all of these congruences using this method may
93
seem very tedious, and it is. Thus, you won't have to solve any systems with a bunch of
congruences on the AMC 10 (or else you're probably doing the problem wrong).
Sample Problem 3. Find the units digit of . How can you
relate this to modular arithmetic.
You probably know how to solve this problem without modular arithmetic. You would
probably start by writing out the sequences of the units digits starting with to get 3,
9, 7, 1. Thus,you would probably think that we can keep on subtracting fours from the
exponent (dividing by , which has a units digit of 1) to get that has the same
units digit , so our answer is 7 (we could also see this as saying that 123 has a
remainder of 3 when divided by 4).
Next, we will relate this to modular arithmetic. Thinking how to express a units digit,
we realize that the units digit of a number is equivalent to that number .
Thinking about what we doing in terms of modular arithmetic again, we see that we
found as the lowest exponent of 3 that is congruent to 1 in
, so thus we factored out from repeatedly. Now, this may not initially, but we
can generalize this beyond the units digit of a number. We call the smallest number
such that the order of n . The concept of the order of a
number in a mod will pop up again in higher level number theory that is not necessary
for the AMC 10, but we will give a brief introduction to these ideas in the exercises.
94
Exercises
4.3.1) Show that, for integers x, y, and z such that x is relatively prime to m and
, .
4.3.2) Use the previous statement to come up with an alternate proof for the Chinese
Remainder Theorem.
95
Solutions
1.1.1) First of all, note because the range is 1, the median must either be just the middle
term, or the two terms that average to it must both be 8. Thus, we know there is an 8 in
the set. However, because the average is greater than 8, we also need numbers greater
than 8, which can only be 9s. We will then break this up into cases based on if n is even
or odd.
First, if n is even, we know that we have two 8s in the middle and can add and 8 and 9
pair by putting one on either side of the two middle 8s when in non-decreasing order.
Thus, we always have two more 8s than 9s, so our average is . Setting this equal
to 8.3, we have . This
works because we assumed a is even, and 10 is indeed even.
Second, if n is odd, we can use similar reasoning to get . Solving, we get a
a=5. This is indeed odd, so thus we get a= 5, 10 as our solutions.
Notice that if you were to try solving the problem with an average of 8.5 you would get
no solutions. The same is true for any average below 8. Try to come up with a
non-algebraic, intuitive explanation for why the average must be between 8 and 8.5,
including 8 and excluding 8.5. Also, note that not every value in this interval can be
formed (can you find an intuitive explanation for this).
1.1.2) First, note that the sum of the elements in the set is , where a is the average
value of the elements in the set. Thus, when we remove an element of value v, the sum
of the remaining elements is reduced to , and the average becomes . Try
letting to see what the average would become in terms of c.
1.2.1) Note that the roots of unity are the solutions to , or . By
Vieta's Formulas, the sum of the roots of this polynomial is equal to the coefficient of the
term divided by the coefficient of . However, because , the term has a
coefficient of and has a coefficient of 1. Thus, the desired sum is .
96
1.2.2) Note that, by Vieta's Formulas, the product of all of possible values of is
negative the value of the constant term divided by the coefficient of the term in the
second equation.
First of all, let's try to find the constant term. By Vieta's Formulas on the first equation,
we get . Next, note that
. Thus, the constant term is
.
.
. However, when
we expand out , we see that it telescopes
nicely because
,
as desired.
1.4.1) We will simply turn the words in the problems into equations and then solve. We
will let the number of cups he has be , the number of ounces of lemonade he had be ,
and the number of cups his cooler be . From our first given we have that .
Second, we have that . Third, we have that . By substituting the first
equation into the second for and multiplying by 2, we get
Thus, using the last equation, we get =
Therefore, our answer is 4.
1.4.2) We will again start by defining variables. Let Mary's rate for the first half of the
track be mph and the distance of the total track be miles. We have the the first half
97
of the track takes 3 minutes or , so . Thus, . Next, we know she
decreased her rate by 1 mph and traveled the second half of the track in 4 minutes or
hours. Thus, we have and . Equation \our two expressions for ,
we get . Therefore, we finally achieve miles.
2.1.1) We will count this quantity by taking the total number of arrangements and
subtracting the arrangements with two blue beads in a row. There are a total of
arrangements and a total of 4 ways to arrange the beads with the 2 blue beads next to
each other. Thus, there are 6 arrangements fitting the desired constraints.
2.2.2) We will use complementary counting. First, we will count the numbers of
multiples of 3 or 8 below 101. We can easily see that there are 33 multiples of 3, 12
multiples of 8, and 4 multiples of 3 and 8. Thus, there are 33+12-4=41 multiples of 3 or 8
below 101 and 100-41=59 positive integers that are not multiples of 3 or 8 below 101.
2.3.1) We will use complementary counting to reduce the amount of casework that we
have to do. If we don't have at least 2 heads come up, then that means that either 0 or 1
head comes up. The probability that exactly 0 heads come up is . The probability that
exactly 1 head will come up is because any of the coins could come up heads. Thus,
the probability that at least 2 heads come up is 1- - = .
2.4.1) By the Binomial Theorem, when we expand we merely get the sum
of the numbers in the nth row of pascal’s triangle.
98
2.4.2) There are several ways to prove this identity, we will just show one. Using
pascal’s triangle, we know that each number is equal to the sum of the two numbers
directly above it. In other words, we get that the sum of the kth and k+1th numbers in
the nth row is equal to the kth number in the n+1th row. However, using the binomial
coefficient definition of Pascal's Triangle, we get can restate the above to get
, which is exactly what we wanted.
This identity that we just proved is called Pascal's Identity.
2.4.3) There are again several ways to prove this. We will only show an algebraic proof
here, but we encourage you to try to find another proof using Pascal's Triangle as we
did in the above solution.
We will perform the following steps using Pascal's Identity (exercise 2) and basic
properties of binomial coefficients:
And thus, we are done.
99
the inner triangle similar to the outside one with a side ratio of 1/2). Thus, the inradius
of the smaller triangle is divided by 4 and multiples by 2, leaving a ratio of the inradius
of the outer triangle to the inradius of the inner triangle to be 2 to 1.
3.2.2) We will simply note that triangle DFC is similar to triangle EGC because angles
CEG and CDF because they intersect the same arc. Thus, we get
.
OR
Similarly to how we proved the last case of the Power of a Point theorem, both
and are equal to the power of point C with respect to circle O.
3.3.1) We know that the centroid is the intersection of the medians. Thus, we simply
need to find the formulas for the medians. The midpoint of AB is (2,3), so we get that
the median to C has the equation y=3x-3. The midpoint of BC is (3,5), so the equation
100
for the median to A is y=3x/2+1/2. These two medians intersect at .
Try to generalize this for a triangle with vertices , , and to see if you
can find a general formula.
of triangle AEB is .
3.4.1) Note that, by symmetry, the center of the cube is also the center of the sphere.
Thus, long diagonal of the cube has length equal to the diameter of the cube, or .
Next, we will find the length of the long diagonal of a cube in terms of the cube's side
length in order to find the cube's side length.
Note that the long diagonal is the hypotenuse of a right triangle with legs equal to the
diagonal of one face and one of the sides. Thus, the square of the length of the long
diagonal is equal to to the sum of the square of the length of one side and the square of
the length of the diagonal of one face. However, by the Pythagorean Theorem, we also
get that the square of the length of the diagonal of one facts is equal to the sum of the
square of the length of two sides of the cube.
3.4.2) Note that the shape formed by connected these centers is just a regular
octahedron. To find the volume of the octahedron, we will split it up into two pyramids
sharing a base that is on the plane parallel to the base of the cube going through the
midpoint of each side perpendicular to the base. We can easily see that this base has an
101
area of half that of the top face of the cube, or an area of 1/2. Because we divided the
octahedron into 2, we can easily see that the height of each pyramid is also 1/2. Thus,
3.5.2) Note that triangle EFB is similar to triangle CEB because angle EFB equals angle
CEB ancf/d angle B equals itself. Thus, we get that EF/FB=CE/EB. Similarly, we see
that triangle EFC is similar to triangle CEB, meaning that CF/EF=CE/EB. Thus, we
have EF/FB=CF/EF or, equivalently, .
3.6.1) First of all, we don't need to worry about the signs of the sines because we are
squaring them. Next, this theorem reminds us of the Pythagorean Theorem. In addition,
we saw in the first diagram that the sine and cosine represent lengths in a right triangle
in the unit circle. However, through this observation we are done! The hypotenuse of
the right triangle seen is the first diagram is the radius of the unit circle, and thus it has
length of 1. Thus, for any angle, we will get the desired equality simply by using the
Pythagorean Theorem and the unit circle definitions for sine and cosine.
3.6.2) See the values that we saw earlier and angles that are related to , we think
equilateral triangles. We will draw the equilateral triangle with side length 2. However,
this doesn't get us a angle or an altitude. Thus, as we know that the altitude from a
vertex of an equilateral triangle is also an angle bisector, we draw an altitude. From
102
there, we have both of the desired angles, so we just need to find CD. We can do this
easily using the Pythagorean Theorem to get . Thus, we get
that and .
4.1.1) We will prove this by showing that applying the Euclidean Algorithm to
is bound to eventually yield . To start, we already know that if we ever get
down to , then g=d because we already proved that the Euclidean Algorithm
will yield the correct greatest common divisor. Next, if at any point is reduced
103
to , then we apply the Euclidean Algorithm again to get to , meaning
r=d, and we have achieved the desired outcome.
We will then assume for the sake of contradiction that we can go without ever reaching
two equal numbers. Thus, if repeatedly applying the Euclidean Algorithm doesn't
reduce us to a case with two equal numbers, one number will always be greater than
the other. Thus, if one number is always greater, we can continually take the lesser
number and subtract it from the greater. This process will clearly never make one of the
two numbers negative because or non-integral because we are taking one integer minus
a smaller integer. However, both numbers are finite, so if we repeat this process an
arbitrarily large number of times we will be subtract at least one from one of the
numbers each time, so we will be subtracting an arbitrarily large amount from their
finite sum. However, this is a contradiction because this would make at least one
number negative, but we said earlier that they must both be positive.
4.1.2) First, note that 13 and 28 are clearly relatively prime. Thus, using the Euclidean
Algorithm will eventually get us to . In other words, we created the 1 by
subtracting two numbers, both of which were formed by subtracting two larger
numbers, and so on until each number was formed from our original two numbers.
Thus, we will essentially start from the bottom of the euclidean algorithm and use
substitutions to get us back to only using the first two numbers. We will start by
computing and showing each step.
104
Now, we could have gone one step further, but we only needed to prove that we
reached a 0 because it helped us prove we would reach the greatest common divisor as
one of the two numbers as well. This is because now that we have a 1, we have all of the
substitutions we need (we aren't trying to find solutions to ).
Continuing on to do the problem, we just substitute as we described before.
Thus, we get x=13, y=-6 as a solution using the euclidean algorithm. You'll see this
more in the Diophantine Equations lesson and you'll how this is useful after the
Modular Arithmetic lessons.
4.2.1) First of all, note that finding an inverse is the same as solving the modular
congruence , or in algebra, . Recall
exercise 2 from the Euclidean Algorithm that we can use the Euclidean Algorithm going
backward to express 1 in the form that gives solutions to this equation when and
are relatively prime as given (exercise 2 was a specific example, but can be easily
general). If you don't understand this exactly, we will fully explain it in the Diophantine
Equations lesson.
105
inverse of 5 in . Thus, multiply both sides of the congruence by 5, we get
.
4.3.2) We wish to prove that there is a solution to the system ,
for m relatively prime to t. We will rewrite the first condition as for some
integer y. Next, note that using the previous exercise, we can see that changes
value whenever y changes in value. In other words, as y ranges over all residues
in , the value of same twice. Thus, we can conclude that the value of
must also range over all residues . Thus, for some value of y, will assume
the value of , thus showing that we do indeed have one solution that is less than
(the greatest residue is t-1). Thus, any remaining solutions must be the one
solution that we found plus a multiple of (as the number we add must be a
multiple of both moduli).
4.3.3) First of all, note that when two numbers that are both relatively prime to the
modulus are multiplied, the result must also be relatively prime to the modulus (if this
isn't clear try writing it out in non-modular notation and use a proof by contradiction).
Thus, if take the set of integers relatively prime to and less than modulus ,multiply
each by a, and take the remainder of each when divided by , we simply get a
rearrangement of the set of numbers relatively prime to and less than m (by exercise 1,
no two distinct integers can produce results congruent when multiplied by a,
but, as we said earlier, each product must also produce a residue relatively prime to the
106