Are there any logical subsets to the 7000 variables? e.g., are they
test items,?
A lot depends on the meanings of the variables.
Art
[log in to unmask]
Social Research Consultants
University Park, MD USA
(301) 864-5570
Ursula Huber wrote:
>Dear all,
>
>I have a question concerning data reduction in an extremely large dataset:
>the data set has more than 7.000 variables (here, categorical variables are
>dummy-coded) and ~30.000 observations. We would like to reduce the dimension
>of the data and thus need to find the ~200 key-variables which explain the
>largest part of the variance in the data.
>
>We already tried factor analysis in spss and clementine -> too many
>variables, the programs crash.
>
>Does anybody have an idea of
>- a very performant statistical software which is able to do factor analysis
>for such a huge data set
>- how we might reduce the dimension of the data without coding the
>categorical variables as indicators
>
>Thank you very much in advance!
>
>Kind regards,
>Ursula
>
>--
>Geschenkt: 3 Monate GMX ProMail + 3 Top-Spielfilme auf DVD
>++ Jetzt kostenlos testen http://www.gmx.net/de/go/mail ++
>
>
>
>
|