Sunteți pe pagina 1din 22

REGRESSION

DEFINITIONS
CORRELATION:
is a statistical method used to determine whether
relationship between variables exists.

 REGRESSION:
is a statistical method used to describe the nature of the
relationship between variables, that is positive or
negative, linear or non-linear.
The purpose of this chapter is to answer
these questions statistically:
Are two or more variables related?
 If so, what is the strength of the relationship?
What type of relationship exists?
What kind of predictions can be made from the
relationship?
Step 1

Scatter plot
A scatter plot is a graph of the ordered
pairs(x, y) of numbers consisting of the
independent variable x, and the
dependent variable y.
A researcher wishes to determine if a person’s age is
related to the number of hours he or she exercises per
week. The data for the sample are shown below.
Draw the scatter plot for the variables.
Age x 18 26 32 38 52 59
Hours y 10 5 2 3 1.5 1
10
8
Hours

6
4
2
Age
0 10 20 40 30 50 60 70
Step 2:

Correlation
Coefficient
The correlation coefficient computed
from the sample data measures the
strength and direction of a linear
relationship between two variables.
The symbol for the sample
correlation coefficient is r.
Range of values for the correlation coefficient
Relationship between the correlation
coefficient and the scatter plot.
Computing r
Example
The director of an alumni association for a small
college wants to determine whether there is any type of
relationship between the amount of an alumnus’s
contribution (in dollars) and the years the alumnus has
been out of school.
The data are shown here.

Years x 1 5 3 10 7 6
Contribution 500 100 300 50 75 80
y
Compute the value of the correlation coefficient.
Years x 1 5 3 10 7 6
Contribution y 500 100 300 50 75 80
 x = 32 y = 1105
 x 2 = 220 y 2 = 364,525
 xy = 3405 n =6
 x = 32  x 2 = 220  xy = 3405
 y = 1105  y 2 = 364,52 n = 6

n   xy  –   x    y 
r =
n
  
x
2

– x  2 

  
n y
2
y  2

 x = 32  x 2 = 220  xy = 3405
 y = 1105  y 2 = 364,52 n = 6

6  3405  –  32  1105 


r=
 2  2
6  220  –  32  
6  364,525  –  1105  

r = – 0.883
Regression
Find the equation of the regression line and find the y
value when x = 70 ºF. Remember that no regression
should be done when r is not significant.
Temperatures ( in. F ) and precipitation (in.)
Avg. daily temp. 86 81 83 89 80 74 64
x
Avg. mo. Precip. 3. 1. 3. 3. 3.7 1. 0.2
y 4 8 5 6 5
 x = 557  y = 17.7
 x 2 = 44,739  xy = 1468.9
 x = 557  y = 17.7
 x 2 = 44,739  xy = 1468.9

a=
 y  
x2 –   x    xy 

2
n x – x 
2

(17.7)(44,739) – (557)(1468.9)
a=
7(44,739) – (557)2

a = – 8.994
 x = 557  y = 17.7
 x 2 = 44,739  xy = 1468.9

n   xy  –   x    y 
b=
 2

n x –  x 2

b= 7(1468.9) – (557)(17.7)

7(44,739) – (557)2

b = 0.1448
 x = 557  y = 17.7 a = – 8.994
 x 2 = 44,739  xy = 1468.9 b = 0.1448

y  = a + bx
y  = – 8.994 + 0.1448 x
y  = – 8.994+ 0.1448(70)
y  = 1.1 inches

S-ar putea să vă placă și