Linear Regression Analysis

In this field the basic methods to carry out regression analysis are studied. The emphasis is on data that are characterised by many variables. The focus is on finding good latent structure that generates the response variables. The latent structure is generated in a way that optimises the prediction aspect of the model. Extensive graphic procedures illustrate the inherent variation of data.


1. Introduction. Ideas of the H-method
2. Latent structures
3. Importance of covariance. The H-principle
4. Linear regression analysis
5. Plots using score vectors
6. Plots showing loading vectors
7. Plots of measures of regression analysis
8. Cross-validation
9. The H-method
10. Quantitative measures for variables
11. Selection/ordering of variables
12. Subset selection of variables
13. Dimension analysis
14. Analysis of residuals
15. Outlier detection
16. Sensitivity analysis
17. Confidence intervals
18. Detection of special features
19. Loading weight vectors
20. Useful measures in linear regression
21. Guidelines for presentation of results