Transcript An Introduction to Logistic Regression

An Introduction to Regression
with Binary Dependent
Variables
Brian Goff
Department of Economics
Western Kentucky University
Introduction and Description





Examples of binary regression
Features of linear probability models
Why use logistic regression?
Interpreting coefficients
Evaluating the performance of the model
Binary Dependent Variables
In many regression settings, the Y variable is (0,1)
A Few Examples:





Consumer chooses brand (1) or not (0);
A quality defect occurs (1) or not (0);
A person is hired (1) or not (0);
Evacuate home during hurricane (1) or not (0);
Other Examples?
Scatterplot of with Y=(0,1):
Y = Hired-Not Hired; X= Experience
1
0
The Linear Probability Model (LPM)
If we estimate the slope using OLS regression:
Hired = α + *Income + e ;

The result is called a “Linear Probability
Model”


The predicted values are probabilities that Y equals
1;
The equation is linear – the slope is constant
Picture of LPM
1
LPM Regression Line
(slope coefficient)
Points on regression line represent predicted probabilities
For Y for each value of X
0
An Example: Loan Approvals
Data:
Dependent Variable: Loaned
1 if Loan Approved, 0 if not Approved by Bank Z
Independent Variables
ROA = net income as % of total assets of applicant;
Debt = debt as % of total assets of applicant;
Officer = 1 if loan handled by loan officer A and 0 if handled
by officer B;
Scatterplot (Loaned – NITA)
LPM Results
Coefficientsa
Model
1
(Cons tant)
nita
tdta
officer
Uns tandardized
Coefficients
B
Std. Error
1.087
.192
.022
.013
-.063
.029
-.279
.138
Standardized
Coefficients
Beta
.237
-.291
-.291
t
5.659
1.655
-2.156
-2.020
Sig.
.000
.105
.036
.049
a. Dependent Variable: loaned
Coefficient on NITA implies 1% increase in ROA increases
Probability of loan by 2.2% (0.022)
LPM Weaknesses

The predicted probabilities can be greater than 1 or
less than 0



The error terms vary based on size of X-variable
(“heteroskedastic”) –


Probabilities, by definition, have max =1; min = 0;
This is not a big issue if they are very close to 0 and 1
There may be models that have lower variance – more
“efficient”
The errors are not normally distributed because Y
takes on only two values


Creates problems for
More of an issue for statistical theorists
Predicted Probabilities in LPM
Loans Model
In loan case, all of the predicted probabilities fall within (0,1) range
Descriptive Statistics
N
predicted_loans
Valid N (lis twis e)
51
51
Minimum
.01273
Maximum
.97245
Mean
.6666667
Std. Deviation
.19034701
(Binary) Logistic Regression or “Logit”



Selects regression coefficient to force predicted
values for Y to be between (0,1)
Produces S-shaped regression predictions rather
than straight line
Selects these coefficient through “Maximum
Likelihood” estimation technique
Picture of Logistic Regression
1
Logistic Regression
(non-linear slope
coefficient)
Points on regression line represent predicted probabilities
For Y for each value of X
0
LPM & Logit Regressions

LPM & Logit Regressions in some cases provide
similar answers
If few “outlying” X-values on upper or lower ends
then LPM model often produces predicted values
within (0,1) band
 In such cases, the non-linear sections of the Logit
regression are not needed
 In such cases, simplicity of LPM may be reason for
use
 See following slide for an illustration

Example where LPM & Logit Results
Similar
LP Model
1
0
LPM & Logit: Loan Case

In Loan example the results are similar:
R-square = 98% for regression of LPM-predicted
probabilities & Logit-predicted probabilities
 Descriptive statistics for both probabilities appear
below:


The main difference is the LPM is max/min closer to 0
and 1
Descriptive Statistics
N
pred_lpm
pred_logit
Valid N (listwise)
51
51
51
Minimum
.01273
.06948
Maximum
.97245
.91364
Mean
.6666667
.6666667
Std. Deviation
.19034701
.19209809
SPSS Logistic Regression Output for
Loan Approval:
Variables in the Equation
Step
a
1
nita
tdta
officer
Constant
B
.108
-.325
-1.455
2.968
S.E.
.070
.180
.767
1.187
Wald
2.393
3.241
3.599
6.248
df
1
1
1
1
Sig.
.122
.072
.058
.012
Exp(B)
1.114
.723
.233
19.443
a. Variable(s ) entered on s tep 1: nita, tdta, officer.
Note: The, instead of t-statistics, “Wald” statistics are used to test whether the
Coefficients differ from zero; the associated p-values (Sig) have the same
Interpretation as in any other regression output
Interpreting Logistic Regression (Logit)
Coefficients

The slope coefficient from a logistic regression
() = the rate of change in the "log odds" of the
event under study as X changes one unit


What in the world does that mean?
We want to know the change in the probability of the
event as X changes

In Logistic Regression, this value changes as X-changes (Sshape instead of linear)
Loan Example:
Effect of NITA on Probability of Loan
NITA coefficient (B) = 0.11
P
(1-P)
B*(P)*(1-P)
Low Probability
0.1
.9
0.009
Medium Probability
0.5
0.5
0.0275
High Probability
.9
.1
0.009
Meaning?
At moderate probabilities (around 0.5) of getting a
loan (corresponds to average NITA of about 5), the
likelihood of getting a loan increases by 2.75% for
each 1% increase in NITA
 This estimate is very close to the LPM estimate of
2.2%
 At the lower and upper extremes (NITA values -/+
teens), the probability changes by only about 0.9%
for a 1 unit increase in NITA

Alternative Methods of
Evaluating Logit Regressions
Statistics for comparing alternative logit
models:



Model Chi-Square
Percent Correct Predictions
Pseudo-R2
Chi-Square Test for Fit
Omnibus Tests of Model Coefficients
Step 1

Step
Block
Model
Chi -square
8.498
8.498
8.498
df
3
3
3
Sig.
.037
.037
.037
The Chi-Square statistic and associated p-value (Sig.)
tests whether the model coefficients as a group equal
zero

Larger Chi-squares and smaller p-values indicate greater
confidence in rejected the null hypothesis of no
Percent Correct Predictions



The "Percent Correct Predictions" statistic assumes that if
the estimated p is greater than or equal to .5 then the event
is expected to occur and not occur otherwise.
By assigning these probabilities 0s and 1s and comparing
these to the actual 0s and 1s, the % correct Yes, % correct
No, and overall % correct scores are calculated.
Note: subgroups for the % correctly predicted is also
important, especially if most of the data are 0s or 1s
Percent Correct Results
35% of loan rejected
cases (0) were correctly
predicted
Classification Tablea
Predicted
loaned
Step 1
Obs erved
loaned
.00
.00
1.00
Overall Percentage
1.00
6
2
11
32
Percentage
Correct
35.3
94.1
74.5
a. The cut value is .500
94% of loan accepted
cases (1) were correctly
predicted
75% of all cases (0,1)
were correctly predicted
Note: The model is much better at predicting loan acceptance than loan rejection – this may serve
as a basis for thinking about additional variables to improve the model
2
R Problems
1
0
Notice that whether using LPM or logit, the predicted values on the regression lines are not near
The actual observations (which are all either 0 or 1). This makes the typical R-square statistic of no
value in assessing how well the model “fits” the data
Pseudo-R2 Values
Model Summary
Step
1
-2 Log
likelihood
56.427a
Cox & Snell
R Square
.153
Nagel kerke
R Square
.213
a. Esti m ati on term inated at i teration num ber 4 because
param eter esti m ates changed by l es s than .001.


There are psuedo-R2 statistics that make adjustment for the
(0,1) nature of the actual data: two are listed above
Their computation is somewhat complicated but yield
measures that vary between 0 and (somewhat close to) 1 much
like the R2 in a LP model.
Appendix: Calculating Effect of Xvariable on Probability of Y

Effect on probability of from 1 unit change in X
= ()*(Probability)*(1-Probability)


Probability changes as the value of X changes
To calculate (1-P) for a given X values:


(1-P) = 1/exp[α + 1*X1 + 2*X2 …]
With multiple X-variables it is common to focus on one at a time and
use average values for all but one