getting a submatrix of all true

[Terry Reedy]

"John Hunter" <jdhunter at ace.bsd.uchicago.edu> wrote in message
news:mailman.1057173484.29754.python-list at python.org...
>
> I have a largish data set (1000 observations x 100 floating point
> variables), and some of the of the data are missing.

All too typical -- missing data are the bane of statistics.

>  I want to try a
> variety of clustering, neural network, etc, algorithms on the data,
> and to keep life simple I want to reduce the dimensions of the
matrix
> so that I have no missing values, since not all the algorithms are
> able to handle them and there is sufficient redundancy in the
> variables that I can afford to lose some.

Statisticians have tried a variety of approaches.  Googling  for '
statistics "missing data" 'will give you some leads if you want.

Terry J. Reedy