Summary
In this chapter, we learned to choose which variables have a greater influence over the prediction variable, Y. Statistical methods to search for these high influence variables include the coefficient of determination and correlation, which indicates a percentage of the relationship of the variables by measuring their variances, and also whether the relationship is direct or inverse depending on the slope sign. t-statistics, f-statistics, and the p-value determine whether we can reject the null hypothesis that the slope is equal to zero.
These tests help to get the statistical confidence of the variables to generate a prediction model while helping to reject the null hypothesis that the slope is equal to zero.
It is important to evaluate all the statistical tests. Sometimes, f-statistics and the p-value indicate that a variable could affect the model with a slope equal to zero. However, we have to understand all the evaluation methods to include the variables that influence...
