Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Interaction term (with empty cells) in ANOVA

From   Martin Korndoerfer <>
Subject   st: Interaction term (with empty cells) in ANOVA
Date   Sat, 20 Apr 2013 13:26:09 +0200

Hi all,

I am facing the following problem:
I want to conduct a simple ANOVA with two fixed factors and their interaction.

hhld = maximum number of children in household (3 levels: 2-4)
position = rank in household (4 levels: 1-4)


anova score_ext position hhld position#hhld

I get the usual output for an ANOVA. However, for the interaction term, three cells are empty (by design). But the output does not change neither if I ?set emptycells drop? nor if I ?set emptycells keep?. Do you know why?

In case of empty cells, I thought a maximum likelihood estimation is the right thing to do. Is Stata perhaps doing this automatically? (Thus, can I use the output?)

Another simple possibility I thought of was the following command:

regress score_ext position hhld position#hhld

Now, Stata recognizes that there are empty (and omitted) cells but there is no longer a coefficient for the whole interaction term. Instead, Stata gives a coefficient for each possible combination of position and hhld (which is not empty or omitted). Is this the right command and is it just not possible to determine a coefficient for the whole interaction term?

Or do I need to create a maximum likelihood estimation by myself? ? In this case, what is the right command to apply?

I appreciate any help.


This message was sent using IMP, the Internet Messaging Program.

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index