Regression Model with Transformations
Regression Model with Transformations
This Demonstration shows the simple linear regression model for various datasets. You can transform the response variable (i.e. the dependent variable), and display the least-squares fit, the coefficient of determination , and estimates of the marginal variance of , as well as the conditional variance assumed to be common across all values of . Box–Cox and Tukey power transformations are available; you can vary the power parameter and display a plot of the effect of various values of on . When the transformation used is either Box–Cox or Tukey's transformation, a plot of versus the parameter is shown.
2
R
y
x
λ
λ
2
R
2
R
λ