[Top][All Lists]

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

Re: ANOVA with empty cells

From: Mike Miller
Subject: Re: ANOVA with empty cells
Date: Mon, 11 Apr 2005 12:54:50 -0500 (CDT)

On Mon, 11 Apr 2005, Isaac Neuhaus wrote:

How does SAS do it?

Source   DF     Type III SS     Mean Square    F Value    Pr > F
R         3     36.30000000     12.10000000      48.40    0.0203
T         1      4.08333333      4.08333333      16.33    0.0561
R*T       1      0.75000000      0.75000000       3.00    0.2254

You'll have to look at their manuals to figure out what they are doing. Note that you have 3 df for R, 1 df for T, but only one df for R*T. It must be fitting a model with a sort of partial interaction effect. In a full model, you would have 3*1 = 3 df for interaction.

Looking more closely, you can only fit an interaction to the part of the model where f1 is 1 or 3. So that must be what it is doing. You can do this in Octave using dummy codes. You need three dummy codes for f1, one for f2, and one for (arbitrary choice here) when f1=1, f2=1. Once you have that done, you can use the '\' operator and other functions to solve this, but I'm sure someone has developed a little regression or ANOVA package to make this easier.


Your data....

    f1   f2   response
S1    1    1    7
S2    2    1    1
S3    3    1    4
S4    4    2    5
S5    3    2    5
S6    1    2    9
S7    3    2    5
S8    1    2    10

Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:
How to fund new projects:
Subscription information:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]