An Iterative Approach to Linear Regression (UMAP)

In this module, a method is given for finding a line or plane fitted to a set of data by using the concept of best fit and letting the computer handle the messy computations involved. By iterative, we mean that we search for the slope of the best fitting line or plane in an interval that becomes successively smaller until we have succeeded in trapping the desired slope.

Table of Contents:

INTRODUCTION

1. BIVARIATE LINEAR REGRESSION
1.1 Using the Computer Program Linreg
1.2 Why Does the T and M (Trap and Magnify) Procedure Work?

2. MULTIVARIATE LINEAR REGRESSION
2.1 Sample Calculation
2.2 Searching for the Best-Fitting Coefficients
2.3 Sample Run of the Computer Program Mulreg
2.4 A Further Comment On the T and M Procedure

3. REFERENCES

4. ANSWERS TO EXERCISES

APPENDIX A. LINREG PROGRAM

APPENDIX B. MULREG PROGRAM