In Regression Forecasting Linearity

In regression forecasting, what are "regression residuals" and how are they utilized in preparing the best forecast? Provide a specific healthcare example of this concept. Ans: Regression residuals are the estimated random error of the random errors of the regression model.Regression residuals are most important thing for forecasting, the best forecasting will happen if the regression residuals are approximately zero.Suppose lungs cancer depend on smoking (not only smoking) , If we try to fit a model for lungs cancer ie lungs cancer= constant+ Parameter* smoking +random error.2. In regression forecasting, what do we mean when we say that there is linearity in a set of data? As a healthcare manager, why is it important for you to test for linearity of data, and how would you go about this task? Ans: In regression forecasting linearity means in the data if we draw the scatter plot of the data points and draw a straight line then every data points lies around the line.Testing for linearity in the data set because of forecasting for future prediction, suppose the data is linear but we fit a quadratic model then these model does not give us better prediction and linear