2. Logistic Regression
• Linear regression :
• Logistic regression
Specify the predictor to include and their term
a type of regression analysis used for predicting the outcome of a
categorical dependent variable (Y)
an approach to model the relationship between a scalar dependent variable
y and one or more explanatory variable denoted X.
Specific model relating the predictors with the outcome .
3. Introduction
Categorical variable = dividing into classes
Example:
Y
Holding/Selling/Buying a stock
Categorical variable with three categories
each of the stock in the dataset (the observations) as
belonging to one three classes
Classifying a new observation, class is unknown, into one of the classes
based on the values of its predictor variable (Classification)
Can be used in data ( the class is known) to find similarities between
observation within each class in term of the predictor variables
(Profiling)
4. Introduction
Applications
1. Classifying customers as returning or returning (classification)
2. Finding factors that differentiate between male and female top executives
(profiling)
3. Predicting the approval or disapproval of a loan based on information such
as scores (classification)
Focus: a binary dependent variable having two possible classes
Popular example of binary response outcomes :
success/failure, yes/no, buy/don’t buy, default/don’t default, and
survive/die
Code the values of binary response Y as 0 and 1
5. Introduction
Steps:
1. Yield Estimates of the probabilities of belonging to each class.
get an estimate of P(Y=1), the probability of belonging to class 1
(which also tells us the probability of belonging to class 0)
2. Use a cutoff value on these probabilities in order to classify each
case in one of the classes.
Example: a cutoff 0.5 ,
case with an estimated probability
P(Y=1) > 0.5 are classified as belonging to class 1
P(Y=1) < 0.5 are classified as belonging to class 0