Count data regression

1. What is the primary purpose of count data regression?

• A) To analyze categorical data

• B) To build models for count data

• C) To compare mean differences

• D) To test for normality

2. Which of the following is a key assumption of Poisson regression?

• A) The variance is greater than the mean

• B) The mean and variance are equal

• C) The data follows a normal distribution

• D) The residuals are independent and identically distributed

3. If the variance is greater than the mean in Poisson regression, what issue arises?

• A) Heteroscedasticity

• B) Overdispersion

• C) Multicollinearity

• D) Endogeneity

4. Which method is commonly used to correct for overdispersion in Poisson regression?

• A) Linear regression

• B) Logistic regression

• C) Negative Binomial regression

• D) Probit regression

5. What is the primary difference between Poisson regression and Negative Binomial regression?

• A) Poisson regression assumes overdispersion, while Negative Binomial does not

• B) Poisson regression assumes mean and variance are equal, while Negative Binomial allows variance to be greater than the mean

• C) Negative Binomial regression is only used for binary data

• D) Poisson regression is non-parametric

6. In Poisson regression, which method is used to check assumptions?

• A) Checking residual plots

• B) Examining the mean and standard deviation of the variable of interest

• C) Performing a t-test

• D) Using ANOVA

7. What should the ratio of deviance to degrees of freedom be in a properly fitted Poisson regression model?

• A) Close to 0

• B) Close to 1

• C) Greater than 2

• D) Exactly equal to the sample size

8. Which of the following is a correction method when Poisson regression assumptions are violated?

• A) Switching to a linear regression model

• B) Applying a log transformation

• C) Using a Negative Binomial regression model

• D) Removing independent variables