Snorke ([info]snorke) rakstīja,
@ 2010-11-20 16:43:00

Previous Entry  Add to memories!  Tell a Friend!  Next Entry
es: The first step would be evaluation of correlation between predictors (deletion of 1 of those predictors who are significantly correlated)


viņš: This not correct!!!! Who told you that you must exclude by a model one predictor due the fact that is highly correlated with another? It is not correct at all!


(Lasīt komentārus)

Nopūsties:

No:
Lietotājvārds:
Parole:
Ievadi te 'qws' (liidzeklis pret spambotiem):
Temats:
Tematā HTML ir aizliegts
  
Ziņa:
Neesi iežurnalējies. Iežurnalēties?