view in publisher's site
- لیست مقالات
Regression diagnostics is the part of regression analysis whose objective is to investigate if the calculated model and the assumptions we made about the data and the model, are consistent with the recorded data. These diagnostics include graphical and numerical tools for checking the adequacy of the assumptions with respect to both the data and the form of the model, detecting extreme points (outliers) that may be dominating the regression and detecting if strong relationships among the independent variables (collinearity) are affecting the results. This works focuses on the multiple linear regression (MLR) model and the ordinary least squares (OLS) estimation of the model parameters. Some regression diagnostics that are used in bilinear (factor-based) regression methods are also commented.