About Mehdi_R

DougWielenga · ‎07-10-2017

A SAS Enterprise Miner license is required to run the NEURAL procedure. Even so, the procedure is not intended to be used directly via programming. There are High-Performance Data Mining procedures such as HPNEURAL that can be run using the SAS Enterprise Miner interface or by writing code to call the HP procedures directly. You can find documentation for the HP procedures at http://support.sas.com/documentation/onlinedoc/miner/index.html Regarding which SAS procedures have which functionality, there are often many different SAS Procedures (or SAS Enterprise Miner nodes) that can be used to accomplish some of the same techniques. The choice among them really depends on your business needs. Feel free to reach out to SAS Technical Support and inquire how to perform a specific set of techniques. Other than that, you can always take time to review the documentation for SAS Enterprise Miner (above) or other SAS Analytical Tools at http://support.sas.com/documentation/ Hope this helps!

Mehdi_R · ‎06-23-2016

Sorry, I copied the other one. Just change "vbar" to "hbar". It doesn't work. Also, data is stored in a secure environment and I can't take it out. But the data layput is similar to this, but 50 chapters not 5: chapter case WgtFreq 1 1 5 1 0 10 2 1 2 2 0 6 3 1 5 3 0 9 4 1 10 4 0 15 5 1 20 5 0 15 ......

Mehdi_R · ‎06-23-2016

Thank you very much. It worked 🙂

Mehdi_R · ‎01-29-2016

I haven't tried it, but I think this procedure does not adjust for correlation.

Mehdi_R · ‎01-25-2016

I got this error when I was using SEX variable whose values are F and M rather than numeric values! Once I chnage SEX to a numeric variable, I didn't get this error anymore...

Mehdi_R · ‎08-21-2015

I see! Thanks for your helpful comments.

SteveDenham · ‎08-21-2015

The following assumes that 'prepost' captures two 'trials' on each subject. The repeated nature can be captured using a G side random statement that models the variance-covariance matrix over time. proc glimmix data = dataset; class knowledge id prepost ; model knowledge (ref = '1') = prepost/ dist = multinomial link = glogit solution ; random intercept /subject = id type = un group = knowledge; random prepost/ type=un subject=id; run; As to why PROC GEE is not running, you need to be on SAS/STAT 14.1 to be able to model a nominal multinomial distribution. See Example 43.6 GEE for Nominal Multinomial Data (http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_gee_examples06.htm) for a worked example. Steve Denham

SteveDenham · ‎08-13-2015

Let it run--you are estimating a lot of parameters and a multinomial model is not easy to fit. I will bet it comes out "Did not converge" at the 20th iteration (the default number of iterations). You will likely have to add an NLOPTIONS statement if that is the case. Steve Denham

Mehdi_R · ‎07-07-2015

Thank you Jacob. It was really helpful:)

Mehdi_R · ‎06-29-2015

Thank you Rick. Actually I got similar results. The relative bias is still between 20 and 70 percent!! (while it should be at most around 5%)

Mehdi_R · ‎06-01-2015

Thanks

Ksharp · ‎05-20-2015

There is a new PROC SSP added into SAS9.4 . Maybe you should take a look at it .

Online Status	Offline
Date Last Visited	‎12-23-2016 12:20 AM

Re: datalabel in hbar statement

Re: datalabel in hbar statement

datalabel in hbar statement

Re: Color of bars change after saving the bar graphs to a pdf

Color of bars change after saving the bar graphs to a pdf

PROC NEURAL

Re: Zero-inflated and correlated count data

Re: Zero-inflated and correlated count data

Zero-inflated and correlated count data

Re: PROC NLMIXED and coding a Likelihook function

Re: PROC SGPLOT - LEGEND AND CURVELABEL TOGETHER

Re: Color of bars change after saving the bar graphs to a pdf

How to build a deep learning model in SAS Enterprise Miner

Re: A problem in PROC NLMIXED

Re: PROC NEURAL

Re: datalabel in hbar statement

Re: Color of bars change after saving the bar graphs to a pdf

Re: Zero-inflated and correlated count data

Re: PROC NLMIXED and coding a Likelihook function

Re: Not a proportion?? PROC GLIMMIX

Re: PROC GEE

Re: multiple (multinomial) outcomes' and repeated measures' correlatio...

Re: Matrix functions in Proc MCMC

Re: REML estimator of variance components in frailty modeling

Re: A problem in PROC NLMIXED

Re: Spatial Statistics in PROC MCMC