topic Re: Factor analysis using maximum likelihood estimation in Statistical Procedures

Base SAS (Geometric Means and Related)

Doug____ — Mon, 08 May 2017 17:13:11 GMT

I've seen similar posts but I am trying to determine if there is a simpler method for this:

1. Create log10 transformed values of x

2. Process through PROC MEANS to obtain mean, sd, median, min max

3. Raise each of the quantities in (2) to 10**(mean), etc. to get "mean", "SD", "min", and "max" values

4. Calculate dispersion factors

-All quantities based on output from Means procedure in step 2

dfplus=(10**mean)+(10**(mean+sd)-10**(mean));

dfminus=(10**mean)-(10**(mean)-10**(mean-sd));

Since I need to calculate grand mean, sd, min, max, dfplus, and dfminus by anaysis category, is there a way I can reduce the number of steps?

Rick_SAS — Mon, 08 May 2017 17:05:06 GMT

I don't understand what your question has to do with the title of this post, "Factor analysis using maximum likelihood estimation."

What are you trying to accomplish? What variables are you using in the factor analysis?

Doug____ — Mon, 08 May 2017 17:06:41 GMT

I chose the incorrect select - can I edit the post so it will show up in the correct location?

Rick_SAS — Mon, 08 May 2017 17:09:27 GMT

Sure. You are welcome to edit the post title, add details, clarify your questions, and so forth.

Ksharp — Tue, 09 May 2017 13:42:53 GMT

Here could get you many statistical estimators .

data class;
 set sashelp.class;
 log_height=log(height);
run;
proc univariate data=class outtable=want noprint;
var log_height;
run;