Ante is a statistical measure used to assess the relative predictive power of a given variable.
It is calculated by comparing the amount of variation in the outcome variable that is explained by the predictor variable to the amount of variation that is explained by all other predictor variables.
How to calculate
-
Calculate the coefficient of determination (R2) for the predictor variable in question. This is done by running a linear regression of the outcome variable on the predictor variable.
-
Calculate the total R2 of the models. This is done by running a linear regression of the outcome variable on all other predictor variables.
-
Subtract the R2 of the predictor variable from the total R2. This is the Ante of the predictor variable.
-
The higher the Ante, the greater the relative predictive power of the predictor variable compared to all other predictor variables.
Examples
- Ante can be used to identify and compare the mean values of two different populations.
- Ante can be used to test for differences in variance between two populations.
- Ante can be used to detect outliers in a data set.
- Ante can be used to compare two distributions to determine if there is a significant difference between them.
- Ante can be used to assess the strength of correlation between two variables.