Logistic Regression Interview Questions & Practice Tests

This page lists down a set of 30 interview questions on Logistic Regression (machine learning/data science) in form of objective questions and also provides links to a set of three practice tests that would help you test / check your knowledge on an ongoing basis. These questions and practice tests are intended to primarily help interns/freshers/beginners to help them brush up their knowledge in logistic regression from time to time.  The following is a list of topics covered on this page.

• Introduction to logistic regression
• Logistic regression examples
• Evaluating performance of logistic regression and related techniques including AIC, deviance, ROC etc.
• Difference between linear and logistic regression

Here is another post on questions and answers related to Logistic regression fundamentals titled, Logistic Regression Quiz Questions and Answers. Here are some of the questions discussed in this post:

1. What are different names/terms used in place of Logistic regression?
2. Define Logistic regression in simple words?
3. Define logistic regression in terms of logit?
4. Define logistic function?
5. What does training a logistic regression model mean?
6. What are different types of logistic regression models?
7. What are the different implementations of Logistic regression in Python Sklearn?
8. What is regularization in Logistic regression and what are its different types?
9. When to use which types of regularization in Logistic regression?

You might also want to check a related post on Logistic regression titled – Training a logistic regression model using Python.

Logistic Regression Practice Tests

This is a set of practice tests (10 questions and answers each) that can be taken to quickly check your concepts on logistic regression. The questions included in these practice tests are listed in a later section.

• Logistic regression practice test – Set 1
• Logistic regression practice test – Set 2
• Logistic regression practice test – Set 3

Logistic Regression Interview Question Set

1. Logistic regression is used to predict _________ valued output?
   • Continuous
   • Categorical
2. How much marks a student can get in a competitive exam based on hours of study can be solved using _________ regression model
   • Multi-linear
   • Logistic
3. Logistic regression is _________ when the observed outcome of the dependent variable can have only two values such as 0 and 1 or success and failure
   • Binomial
   • Multinomial
   • Ordinal
4. Whether a student will pass or fail in the competitive exam based on hours of study can be solved using _________ regression model
   • Multi-linear
   • Logistic
5. ________ regression can be termed as a special case of _________ regression when the outcome variable is categorical
   • Logistic, Linear
   • Linear, Logistic
6. In logistic regression, the goal is to predict _________
   • Actual value of outcome dependent variable
   • Odds of outcome dependent variable
7. Which of the following can be used to evaluate the performance of the logistic regression model?
   • Adjusted R-Squared
   • AIC
8. Which of the following is link function in logistic regression
   • Identity
   • Logit
9. Logistic regression is _________ when the observed outcome of the dependent variable can have multiple possible types
   • Binomial
   • Multinomial
   • Ordinal
10. In logistic regression, the following technique is used to measure the goodness of the fit
    • Sum of squares calculations
    • Deviance calculations
11. Which of the following can be used to evaluate the performance of the logistic regression model?
    • AIC
    • Null and Residual Deviance
    • Both of the above
    • None of the above
12. Given two models with different AIC values, which one would be the preferred model?
    • One with a higher AIC value
    • One with a lower AIC value
13. Deviance is a measure of difference between a _______ model and the _________ model
    • saturated, fitted
    • Fitted, saturated
14. Logistic regression is _________ when the observed outcome of dependent variable are ordered
    • Binomial
    • Multinomial
    • Ordinal
15. Logit transformation is log of ___________
    • Odds of the event happening for different levels of each independent variable
    • The ratio of odds of the event happening for different levels of each independent variable
16. Logistic function is _________
    • Dependent variable equalling a given case
    • Probability that dependent variable equals a case
17. Deviance is is a function of ________
    • Exponential function of likelihood ratio
    • Logarithmic function of likelihood ratio
18. The odds of the dependent variable equaling a case (given some linear combination x of the predictors) is equivalent to _______
    • Log function of the linear regression expression
    • Exponential function of the linear regression function
19. Regression coefficients in logistic regression are estimated using ________
    • Ordinary least squares method
    • Maximum likelihood estimation method
20. _________ is analogous to __________ in linear regression
    • Sum of squares calculations, deviance
    • Deviance, the sum of squares calculations
21. Deviance can be shown to follow __________
    • t-distribution
    • F-distribution
    • Chi-square distribution
    • None of the above
22. ______ value of deviance represents the better fit of the model
    • Higher
    • Lower
23. If the model deviance is significantly ________ than the null deviance then one can conclude that the predictor or set of predictors significantly improved model fit
    • Smaller
    • Larger
24. Which of the following is analogous to R-Squared for logistic regression
    • Likelihood ration R-squared
    • McFadden R-squared
    • Cox and Snell R-Squared
    • All of the above
25. Estimation in logistic regression chooses the parameters that ___________ the likelihood of observing the sample values
    • Minimizes
    • Maximizes
26. Which of the following tests can be used to assess whether the logistic regression model is well calibrated
    • Hosmer-Lemeshow test
    • ROC Curve
    • Both of the above
27. ROC related with ROC curve stands for _______
    • Regression Optimization Characteristic
    • Regression Operating Characteristic
    • Receiver Operating Characteristic
28. Which of the following is used to identify the best threshold for separating positive and negative classes
    • Hosmer-Lemeshow test
    • ROC Curve
    • Both of the above
29. ROC curve is a plot of __________ vs ___________
    • Sensitivity, 1-specificity
    • 1-specificity, Sensitivity
30. ______ the value of AUC, better is the prediction power of the model
    • Lower
    • Higher