MTH-416, REGRESSION ANALYSIS Lecture Notes - Lecture 52: Three Steps, Row And Column Vectors, Variance

• School: Indian Institute of Technology - IIT Kanpur
• Department: Mathematics
• Course: MTH-416, REGRESSION ANALYSIS

Document Summary

Usually, it is difficult to compare the regression coefficients because the magnitude of j reflects the units of measurement of thj explanatory variable. jx for example, in the following fitted regression model y. 1x in litres and y is measured in litres, X when another variable is held fixed explanatory variables is identical. One litre change in either produces the same change is y . Sometimes it is helpful to work with scaled explanatory variables and study variable that produces dimensionless regression coefficients. These dimensionless regression coefficients are called as standardized regression coefficients. There are two popular approaches for scaling, which gives standardized regression coefficients. We discuss them as follows: unit normal scaling: Employ unit normal scaling to each explanatory variable and study variable. 1,2,, k i x ij x j z ij y i y i j s s y y n where. 2 y are the sample variances of thj explanatory variable and study variable, respectively.