SAS Support Communities

davis · ‎09-22-2021

Good afternoon, I have a question, and was wondering if someone could either lend a hand or point me in the right direction. I conducted a multinomial run using proc logistic. One output is the predicted probability for each outcome in the following form: Pred_A=exb_A /(1+exb_A+exb_B+exb_C+exb_D); Pred_B=exb_B /(1+exb_A+exb_B+exb_C+exb_D); Pred_C=exb_C /(1+exb_A+exb_B+exb_C+exb_D); Pred_D=exb_D /(1+exb_A+exb_B+exb_C+exb_D); Pred_E=1 /(1+exb_A+exb_B+exb_C+exb_D); where Pred_E = 1/(1+Pred_A+Pred_B+Pred_C+Pred_D) and produced automatically by proc logistic. Does HP Logistic have a similar feature for a multinomial run? Thanks.

davis · ‎09-17-2021

CONTROLLED Thx! I will explore.

davis · ‎09-17-2021

Good afternoon, I have a quick question regarding the aforementioned subject, and was wondering if someone could lend a hand. To no avail, I have been exploring proc HP Logistic (SAS EG) for a multinomial case, and unsure where I a making syntax errors. If someone could share or point me to sample code that works for proc hplogistic for a multinomial case, it would be sincerely appreciated. Thanks.

davis · ‎01-26-2021

CONTROLLED Thank you. proc logistic/effect statement is where I started. runs fine, but relatively slower than proc hplogistic.

davis · ‎01-25-2021

I am new to hplogistic, and was wondering if it possesses the capability to specify splines for continuous covariates? If so, pointing me in the right direction, providing a reference, or sending a link would be much appreciated. Thanks.

davis · ‎06-03-2020

Good morning, I have a question, and was wondering if someone could lend a hand. I am running a competing risk model (default risk and prepayment risk) using proc logistic via SAS EG (7.15 7.100.5.5850 64-bit). I have one set of covariates, or right-hand side variables, for both risks. However, I need some flexibility. That is, I would like to restrict (i.e., omit) one covariate from this set of covariates in predicting one of the risks. For instance, the covariate 'incentive to refinance a mortgage' is driven by mortgage rates and does not help explain default. As a result, I would like to 'turn off' the 'incentive to refinance a mortgage' covariate in predicting mortgage default. Currently, I am exploring the restrict statement, but I have not seen as of yet, an application of this statement applied to proc logistic. Any assistance would be much appreciated.

davis · ‎03-24-2020

Good morning, I am interested in exploring out-of-sample testing using SAS EG proc adaptivereg (SAS EG, 7.15 (7.100.5.5850, 64-bit) and the partition statement. However, I am running into a little difficulty. My input data set is a "skinny" and unbalanced panel (i.e., time series, cross section), and the partition statement seems to operate on an observation, or row, as opposed to a group. Applying the partition statement to the panel will ignore, it seems, the time dimension, which needs to be preserved. If my understanding is clear, any thoughts on how to get the partition statement to operate on a group (e.g., unique member id) as opposed to an observation or on any other ways to potentially navigate this obstacle? Thanks.

davis · ‎01-28-2020

Good afternoon, I am trying to modify the structure of a data set and unsure how. Data Set I have: Census_st_cde 01/01/1990 02/01/1990 03/01/1990 04/01/1990 AL 6.72 6.71 6.68 6.64 AK 7.17 7.21 7.22 7.21 Data Set I need: Census_st_cde 01/01/1990 AL 6.72 01/01/1990 AK 7.17 02/01/1990 AL 6.71 02/01/1990 AK 7.21 03/01/1990 AL 6.68 03/01/1990 AK 7.22 04/01/1990 AL 6.64 04/01/1990 AK 7.21 Any thoughts would help. Thanks.

davis · ‎12-06-2019

Thanks! I received an email a little ways back indicating that you had planned to retire. Have I been misinformed? I hope so!

davis · ‎12-05-2019

Good afternoon, I need a hand with the appropriate interpretation of the components panel in SAS/STAT 15.1 User's Guide The ADAPTIVEREG Procedure. In Chapter 25 of this Guide on page 1,000, Figure 25.8 titled 'Component Panel 1' displays four component panels (see attached GIF File). The accompanying text of figure is as follows: "The component panels (Figure 25.8 and Figure 25.9) display the fitted functional components against their forming variables." Questions: 1. What is the actual interpretation of '... fitted functional component...'? 2. What is the actual interpretation of '... forming variables ...'? 3. One panel in Figure 25.8 describes a decreasing bivariate relationship between f(horsepower) and horsepower and a kink in this relationship where horsepower=100. I assume the kink represents a slope change at horsepower=100. However, (3a) what is the appropriate interpretation of the downward bivariate relationship between the fitted functional component of horsepower and horsepower and (3b) what does this relationship imply? Assistance would be most appreciated. Thanks.

davis · ‎08-03-2019

Thanks. The coding error has been identified. Row values were not totaled. So, the denominator used in the calculation was not the 'total'. The calculation makes sense now that this adjustment was made. Manual checks. What would we do without them.

davis · ‎07-30-2019

Good afternoon, I have a quick question, and was wondering if someone in the community could help. I am running proc tabulate on some data. To be sure the code ran as intended, I manually computed a few cells. Attached is a snapshot from proc tabulate, which reports a 3.69 row percent, a numerator of 642,583 and denominator of 16,780,658. As a hedge, I manually computed 642,583 / 16,780,658 and came up with 3.83%. I ASSUME the following code is incorrect. Thoughts on the apparent discrepancy? Thanks. ----------- proc tabulate data=DSN format=comma20.2; CLASS STATE / ORDER=UNFORMATTED MISSING; CLASS Flag1 / ORDER=UNFORMATTED MISSING; CLASS D120_flag / ORDER=UNFORMATTED MISSING; TABLE /* Row Dimension */ STATE*( D120_flag*( N RowPctN) ALL={LABEL="Total (ALL)"}*N) ALL={LABEL="Total (ALL)"}*( N RowPctN D120_flag*( N ColPctN)), /* Column Dimension */ Flag1; ; * freq cur_upb_amt; run; ;

davis · ‎04-22-2019

Thank you.

davis · ‎04-22-2019

Thanks. I understand PROC LOGISTIC is an alternate path, at least theoretically. However, I am trying to assess the feasibility of whether either PROC GAM, PROC ADAPTIVEREG or both can handle a multinomial nominal dependent variable.

davis · ‎04-19-2019

Thanks for your reply. You are correct. I am searching for either a nonparametric or semi-parametric fit, where the dependent variable can be specified as a multinomial. I will try your suggestion.

Online Status	Offline
Date Last Visited	‎04-28-2024 03:13 AM

SAS Support Communities

Proc Logistic v. HP Logistic Certain Output Predicted Probabilities Mu...

Re: Sample Code for Proc HP Logistic (SAS EG) for a Multinomial Case

Sample Code for Proc HP Logistic (SAS EG) for a Multinomial Case

Re: Proc Hplogistic and Splines

Proc Hplogistic and Splines

RESTRICT STATEMENT FOR PROC LOGISTIC

Proc Adaptivereg, Partition Statement, Out-of-Sample Testing

REARRANGE DATA

Re: Components Panel in SAS Proc Adaptivereg (SAS EG 7.13 HF4)

Components Panel in SAS Proc Adaptivereg (SAS EG 7.13 HF4)

What does "GCV R-Square" from PROC ADAPTIVEREG mean?

Proc Logistic v. HP Logistic Certain Output Predicted Probabilities Mu...

Re: Sample Code for Proc HP Logistic (SAS EG) for a Multinomial Case

Sample Code for Proc HP Logistic (SAS EG) for a Multinomial Case

Re: Proc Hplogistic and Splines

Proc Hplogistic and Splines

RESTRICT STATEMENT FOR PROC LOGISTIC

Proc Adaptivereg, Partition Statement, Out-of-Sample Testing

REARRANGE DATA

Re: Components Panel in SAS Proc Adaptivereg (SAS EG 7.13 HF4)

Components Panel in SAS Proc Adaptivereg (SAS EG 7.13 HF4)

Re: Apparent Discrepancy Between Proc Tabulate and a Manual Calculatio...

Apparent Discrepancy Between Proc Tabulate and a Manual Calculation (S...

Re: PROC GAM, PROC ADAPTIVEREG, MULTINOMIAL DEPENDENT VARIABLE

Re: PROC GAM, PROC ADAPTIVEREG, MULTINOMIAL DEPENDENT VARIABLE

Re: PROC GAM, PROC ADAPTIVEREG, MULTINOMIAL DEPENDENT VARIABLE

SAS Analytics Explorers

Follow Us

What is...