LernStats
(Chapter 6 - Page 1 / 7)
Introduction - linear regression analysis
The goal of regression analysis is to find a straight line that optimally describes two variables, usually represented by dots in a scatterplot.
This line is calculated via the method of least squares. This means that the sum of all differences between the estimated values and the real values (=residuals) are zero.
The regression line can be calculated via a general, linear regression equation with the regression coefficients a and b:
With the regression equation specific values of a variable Y (criterium) can be estimated or predicted based on the knowledge of specific values of a variable X (predictor).
This is only possible, if there is a correlation between the two metric variables X and Y.
Regression analysis is used when:
- timely later information is to be predicted
on the basis on existing information
- hard to obtain information is to be
estimated on the basis of easily obtained information
|