Question

Subject Name: Data Management for Analytics
Software used: SAS

Choose the best single option.

Q1. Which of the following statements is false regarding categorical input variables and cardinality?

a. Cardinality is the number of distinct levels in a categorical variable.
b. Categorical variables with high cardinality can lead to overfitting in a predictive model.
c. Calculating cardinality is an important exploratory tool before transforming the categorical input.
d. Categorical variables with a low cardinality ratio have more distinct levels than variables with a high cardinality ratio.
e. All are false.

Answer 1

In a dataset Cardinality is the number of distinct levels in a categorical variable

a) Statement 1 is true

Since Cardinality is the number of distinct levels in a categorical variable

b) Statement 2 is true i.e. Categorical variables with high cardinality can lead to overfitting in a predictive model.

c) Statement 3 is true i.e Calculating cardinality is an important exploratory tool before transforming the categorical input

d) Statement 4 is False

Since Categorical variables with a low cardinality ratio have less distinct levels than variables with a high cardinality ratio.

Hence, option d is correct.