Module 6B Regression - Modelling Possibilities
Module 6B Regression - Modelling Possibilities
MODELING POSSIBILITIES
Learning Objective:
At the end of the lesson, the student should be able to:
• Use different types of explanatory variables for inclusion in the regression
equation.
• Interpret the regression results with these types of predictor variables.
• Familiarize oneself with the different types of nonlinear regression.
• Identify when binary logistic regression is appropriate.
• Formulate the binary logistic regression from the software output.
• Predict using the binary logistic regression equation.
Modeling Possibilities
Business Analytics: Data analysis and decision-making by Aljbright and Winston, 2020
DUMMY VARIABLES
Some potential variables are categorical and cannot be measured on a
quantitative scale. To include them in a regression equation, the trick is
to use dummy variables, also called indicators or 0-1 variables.
If a dummy variable for a given category equals 1, the observation is in
that category; if it equals 0, the observation is not in that category.
For example, if gender is a predictor variable, a single dummy variable is
required. Code Gender as 1 for females and 0 for males (or vice versa).
Note:
The number of dummy variables is equal to the number of
categories of the original variable minus 1.
Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0%
Intercept 30.621 4.819 6.355 4.2621E-06 20.536 40.707 20.536 40.707
Male 3.082 3.342 0.922 0.36797016 -3.913 10.078 -3.913 10.078
MBA 3.300 6.049 0.546 0.59171448 -9.360 15.961 -9.360 15.961
Years of Experience 2.890 0.251 11.524 5.1202E-10 2.365 3.415 2.365 3.415
MBA x Years 1.321 0.778 1.698 0.10583088 -0.307 2.950 -0.307 2.950
Over 50 -7.306 6.093 -1.199 0.24527495 -20.059 5.448 -20.059 5.448
Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0%
Intercept 30.621 4.819 6.355 4.2621E-06 20.536 40.707 20.536 40.707
Male 3.082 3.342 0.922 0.36797016 -3.913 10.078 -3.913 10.078
MBA 3.300 6.049 0.546 0.59171448 -9.360 15.961 -9.360 15.961
Years of Experience 2.890 0.251 11.524 5.1202E-10 2.365 3.415 2.365 3.415
MBA x Years 1.321 0.778 1.698 0.10583088 -0.307 2.950 -0.307 2.950
Over 50 -7.306 6.093 -1.199 0.24527495 -20.059 5.448 -20.059 5.448
Step 2: Open in
JASP and run Logistic
Regression.
We can use the fitted model to estimate the probability that a given customer will have a broker account. For example, if a customer has a savings
account balance of 6000 (thou), the predicted probability is
exp(−4.001 + .001 𝑆𝑎𝑣𝑖𝑛𝑔𝑠) exp(−4.001 + .001 𝑥 6000) 7.3817
𝑦ො = 𝑃 𝐵𝑟𝑜𝑘𝑒𝑟 = = = = 0.8807
1 + exp(−4.001 + .001 𝑆𝑎𝑣𝑖𝑛𝑔𝑠) 1 + exp(−4.001 + .001 𝑥 6000) 8.3817
On the other hand, if a customer has a savings account balance of 4000 (thou), the predicted probability is