Search results

  1. A

    Comparing nested models using F test

    Hi all, I want to compare two models to see if they are significantly different. Total number of obs = 981 Model 1: y = b0 + b1x1 + b2x2 + b3x2 Model 2: y = b0 + b1x1 + b2x2 .... + b5x5 How do I compare the two regressions? Is using the difference in F-values for both models, with df1 = 2...
  2. A

    Weighted linear regression assumptions

    Thank you for the response. There are about 200 or so models, so looking at qqplots is not the most efficient way, so I was looking for some formal tests. Each model has anywhere from 100 - 1000 obs.
  3. A

    Weighted linear regression assumptions

    Hi all, question on linear regression with weights - when testing residual assumptions (homoscedasticity and normality), do I need to weight the residuals with the same weight? I am using proc reg "spec" option to check for homoscedasticity and proc univariate "normal" option to check for...
  4. A

    Does SAS get confused by very small numbers?

    I am trying to calculate a population stability index using a dataset (same can be replicated on a made up dataset). When I I have >1 numbers, PSI is calculated correctly. However, when I scale every value down equally by 100k, the reference dataset's bins are still correct, but the second...