Premium
ESTIMATING A NON‐GAUSSIAN REGRESSION MODEL WITH MULTICOLLINEARITY
Author(s) -
And R. B. Cunningham,
Heathcote C. R.
Publication year - 1989
Publication title -
australian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 0004-9581
DOI - 10.1111/j.1467-842x.1989.tb00494.x
Subject(s) - multicollinearity , heteroscedasticity , econometrics , context (archaeology) , variance inflation factor , skewness , statistics , principal component regression , mathematics , linear model , linear regression , regression analysis , computer science , geography , archaeology
summary The paper discusses practical issues that arose in estimating a Linear model subject to heteroscedasticity, skewness and multicollinearity. The context of the problem required that linearity of the model as originally formulated be preserved. A seemingly satisfactory fit was obtained using a Gamma model and principal components.