Stepwisefit number of terms included (easy question)
in [Matlab]
|
From: condor on 2 Jun 2010 22:30 "Tom Lane" <tlaneATmathworksDOTcom(a)nospam.com> wrote in message <hu64rn$5oo$1(a)fred.mathworks.com>... > >> When stepwisefit has 3 predictors, it's not possible to compute the > >> significance of a 4th one > > > I don't understand why... Because of 5 observations I should have a > > maximum number of predictors of 4 + the constant term, isn't it? > > As written above, the r-squared is almost alway near to 1 (from 0.995 to > > 1). The number of predictors used vary from 1 to 3 (+constat term), what > > is strange is that on 20,000 regressions I did, I haven't found one with 4 > > predictors (+constat term)? > > In the process of deciding whether to add a predictor, we (the stepwisefit > function) computes F statistics that compare: > > 1. The reduction in error sum of squares that results from adding the > predictor. > > 2. The remaining error sum of squares after adding the predictor. > > If adding a new predictor will saturate the model, we're guaranteed that #1 > is 100% of the error sum of squares, and #2 is zero. So we can't > meaningfully test one against the other. > > -- Tom > SO you mean that if n is the number of observations and k the number of factors we should always have n>k (never n>=k) right? And this because we heed to have at least one degree of freedom in order to test a model against another?
From: Tom Lane on 3 Jun 2010 11:46 > SO you mean that if n is the number of observations and k the number of > factors we should always have n>k (never n>=k) right? And this because we > heed to have at least one degree of freedom in order to test a model > against another? That is correct, stepwisefit can only produce models with k<n, because it needs at least on degree of freedom for error for testing. -- Tom
From: condor on 4 Jun 2010 06:20
"Tom Lane" <tlaneATmathworksDOTcom(a)nospam.com> wrote in message <hu8ip9$8tv$1(a)fred.mathworks.com>... > > SO you mean that if n is the number of observations and k the number of > > factors we should always have n>k (never n>=k) right? And this because we > > heed to have at least one degree of freedom in order to test a model > > against another? > > That is correct, stepwisefit can only produce models with k<n, because it > needs at least on degree of freedom for error for testing. > > -- Tom > Wondeful, thanks very much |