13 2 points It's important to conduct residual analysis before you begin the model building process so you will know if you're violating any of the assumptions of least squares regression True False 14 2 points Which of the following is most often a problem when we are doing time series forecasting: multicollinearity misspecification heteroscedasticity autocorrelation 15 2 points Some residuals will be smaller or larger than others, but every observation in a data set will have a residual value. True False C