Documente Academic
Documente Profesional
Documente Cultură
Variate is a linear combination of variables with empirical weights. It is the building block of
multivariate.
Multivariate Techniques:
1. Multiple Linear Regression
2. Logistic Regression (Targeting)
3. Factor Analysis (Data Reduction)
4. Cluster Analysis (Market Segmentation)
5. Conjoint Analysis
6. Discriminant Analysis
7. Multi-Dimensional Scaling (Brand Positioning)
Discriminant Analysis
Only one dependent variable – nonmetric
Can be one or more independent variable – must be metric
Conjoint Analysis
Only one dependent variable – nonmetric or metric
One or more independent variables – must be nonmetric
Can be used for new product/service development
Factor Analysis
Factor means common underline dimensions.
It is just simplifying data into different factors.
The score of different dimensions in one factor is summated score.
For every factor, we can have summated score.
We have to then only deal with summated score per factor.
Factor analysis is performed to achieve Parsimony by reducing dimensions with
minimum variables. (Parsimony: Avoiding variables indiscriminately)
Specification Error means omitting a critical predictor variable which are important
for Factor analysis.
Overfitting and Multicollinearity
Overfitting is defined as sample conclusions which are not generalizable for
population.
Multicollinearity is main villain for Multiple Regression.
Cluster Analysis
Cluster Analysis is an analytical technique for developing meaningful subgroups of
individuals.
Segmentation is done in different segments or subgroups.
Session 3 & Session 4
Multi-Dimension Scaling
It is used for Brand Positioning
Session 8
Variance Inflation Factor when using SPSS for each and every independent variable.
If VIF> 5, then it causes multi-collinearity for any independent variable.
If the chocolates are kept at 5ft height, the sales will decrease as compared to 6ft height.
If the chocolates are kept at 6ft height, the sales will increase to 20.25 units as compared
to kept at 7ft height.
Session 9
Durbin Watson is used to check Auto-Correlation in SPSS.
Durbin Watson should be between 1 and 3.
Durbin Watson if exactly at 2 is very good, then there is no auto-correlation.
VIF is used to check multicollinearity, it should be below 5 in SPSS.
Session 10
Next week starts with Logistic regression
Session 11
Formula for likelihood = if (subscription cell=1, probability cell, 1-probability cell)
Session 12
Logistic Regression= Logarithm of Odds
Odds = Probability of happening of an event/Probability of not happening of an event
Odds = Prob/(1-Prob)
Prob = Odds/(1+Odds)
Session 13
Discriminant Analysis using SPSS.
Session 14
Conjoint Analysis
Session 15
Conjoint Analysis
Logistic
MDS(MultiDimensionalScaling)
Conjoint
Discriminant analysis