We've got some data containing two variables, where x is the predictor and y is the response variable. We make a model of the form of: y = α + β ⋅ x + ϵThen we see that in the residual plot (residuals vs. y ^ ) the variance is increasing as y ^ . We then decide to transform our model to a logarithmic form, i.e.: l o g ( y ) = α + β ⋅ x + ϵWhen performing a residual plot analysis, do we plot residuals vs. l o g ( y ) ^ or y ^ ?

Question

We&#039;ve got some data containing two variables, where   x is the predictor and   y is the response variable. We make a model of the form of:  y  =  α  +  β  ⋅  x  +  ϵThen we see that in the residual plot (residuals vs.       y    ^  ) the variance is increasing as       y    ^  . We then decide to transform our model to a logarithmic form, i.e.:  l  o  g  (  y  )  =  α  +  β  ⋅  x  +  ϵWhen performing a residual plot analysis, do we plot residuals vs.             l      o      g      (      y      )        ^   or       y    ^  ?

Jaydan Gilbert · Accepted Answer

You should plot them against the   log  ⁡  (  y  ) as these are the residuals that need to be tested for the logarithmic form

orkesruim40 · Accepted Answer

Residuals are differences between what is what is observed and what is predicted by the regression equation. So if you&#039;re regressing log(y) on x, then you&#039;d use predicted and observed log(y) to find the residuals.

Open question

Answer & Explanation

New Questions in High school statistics