# Enter your data for Random Forests® Classification

Predictive Analytics Module > Random Forests® Classification
###### Note

Select the option that best describes your data.

## Binary response

Complete the following steps if the categorical response data has two categories, such as pass and fail.

1. From the drop-down list, select Binary response.
2. In Response, enter the column that contains the binary response. Values may be numeric or text.
3. In Response event, select which event the analysis will describe. By default, the second response level is selected as the response event. Changing the response event does not affect the model, but it can make the results more meaningful.
4. In Continuous predictors, enter the continuous variables that may explain or predict changes in the response. Continuous predictors must use numeric values.
5. In Categorical predictors, enter the categorical variables that may explain or predict changes in the response. Categorical predictors may use text or numeric values.
In this worksheet, Bought is the binary response that indicates whether a consumer purchased a new brand of cereal. The response event is Yes. Income and Children are continuous predictors. Store and ViewAd are categorical predictors.

The first row in the worksheet shows that the consumer bought the new brand of cereal. This consumer has an income of \$37000, shops at store A, has 1 child, and viewed the advertisement for the cereal.

C1-T C2 C3-T C4 C5-T
Yes \$37,000 A 1 Yes
No \$47,000 A 3 No
Yes \$34,000 A 0 No
Yes \$58,000 B 0 No

## Multinomial response

Complete the following steps if the categorical response variable has three or more categories, such as sedan, truck, and SUV.

1. From the drop-down list, select Multinomial response.
2. In Response, enter the column that contains the categorical response. Values may be numeric or text.
3. In Continuous predictors, enter the continuous variables that may explain or predict changes in the response. Continuous predictors must use numeric values.
4. In Categorical predictors, enter the categorical variables that may explain or predict changes in the response. Categorical predictors may use text or numeric values.
In this worksheet, Target is the multinomial response that indicates whether a loan applicant is a low, medium, or high risk. Income and NumCards are continuous predictors. Marital is a categorical predictor.

The first row in the worksheet shows that a low-risk applicant has an income of 2399, has 3 credit cards, and is single.

C1-T C2 C3 C4-T
Target Income NumCards Marital
Low 2399 3 Single
Medium 2915 5 Single
Medium 3100 0 Married
High 1500 8 Married