Forecasting based on multiple regression

In the first recipe of this chapter, Basic forecasting and statistical inference, we learned how to perform forecasting with simple linear regression. In this recipe, we will learn how to perform forecasting based on multiple regression. Multiple regression is a type of forecasting procedure in which we use more than one variable to predict the outcome variable that we are interested in. In this recipe, our goal is to predict the level of cortisol at the highest effort during the physical exercise, based on cortisol level at rest and cortisol level at the beginning of the test. In the dataset that we are going to use, we have some respondents with missing data for cortisol level during physical exertion. Our aim is to use the result of our regression analysis to approximate cortisol level for those respondents and use these predicted values...