Zur Startseite  
 
  LernStats    Glossary    Textbook    Imprint    
backIndexnext
 



Glossary

Clicking on the glossary keywords on the right will display a definition of the word in this box.

Please note that the glossary definitions at present are available in german language only.


Interdisziplinäres Zentrum für Hochschuldidaktik - IZHD, Hamburg

IEE
   / Home / LernStats / Simple Regression / Introduction - Linear Regression Analysis


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


 
  LernStats    Glossary    Textbook    Imprint    
backIndexnext