[R] Hypothesis Testing using Wald Criterion for two regression models with dummy variables

[meredith]

Peter-
  Maybe I have not articulately my problem clearly, I have had local help
with the statistical part just trying to figure out how to correctly program
this test.  For clarity's sake, I have months worth of data, I want to
potentially combine those months into four, shall we say seasons, that have
close to the same behaviour.  Therefore to do this, I am trying a monthly
moving window to categorize these seasons.  After talking to a couple water
resources statisician's we decided the way to test if the months are
different is through the use of hypothesis testing and a dummy variable. So
I have one regression, Model A, that includes a zero in the dummy spot with
the two months of data combined, then I have another regression, Model B,
that includes the interaction term for the changes between the months (the
intercept changes, using a 0 or 1 dummy variable). Now we discussed running
a Wald testing, Chi squared, to test to see if the interaction term is of
importance probability wise, can I do this utilizing anova? Does this make
more sense?  Then I will run another set of restricted and unrestricted
models to account for potential differences in the mean (i.e the slope).
Does this explain my problem better?

Meredith

--
View this message in context: http://r.789695.n4.nabble.com/Hypothesis-Testing-using-Wald-Criterion-for-two-regression-models-with-dummy-variables-tp4601582p4603260.html
Sent from the R help mailing list archive at Nabble.com.