cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
Pre-app
‎02-16-2018
  • 5 Posts
  • 1 Likes Given
  • 0 Solutions
  • 0 Likes Received

How to know about the # of variables/parameters used by each model at ...

by Calcite | Level 5 in SAS Data Science
‎02-14-2018 04:02 PM
‎02-14-2018 04:02 PM
Hello,   I am using SAS EM 14.2.   There are various models connecting to node [Model Comparison] for assessment. Is there a way to see how many variables/parameters used by each model at node [Model Comparison]?   Thanks. ... View more

Re: Help: why there are different results (Logistic Regression) from P...

by Calcite | Level 5 in SAS Data Science
‎11-27-2017 11:33 AM
‎11-27-2017 11:33 AM
Thank you Astounding.   The 2 procedures use the same dataset for comparison. No sampling is applied.   For example, * For PROC LOGISTIC (also can use node [SAS Code] in SAS EM which would generate the same results as SAS EG), get:                         Analysis of Maximum Likelihood Estimates                                                        Standard          Wald Parameter                         DF    Estimate       Error    Chi-Square    Pr > ChiSq   Intercept                          1     -4.2280      0.1005     1768.1503        <.0001 F_HV                               1     -0.4514      0.0447      102.1930        <.0001 F_REV                              1      1.1634      0.0399      851.3617        <.0001 F_ACTIVE0                          1      0.6207      0.0395      246.4798        <.0001 (The results for other 15 parameters almost match those from PROC DMREG.)                                     Odds Ratio Estimates                                                           Point          95% Wald Effect                                               Estimate      Confidence Limits F_HV                                                  0.637       0.583      0.695 F_REV                                                 3.201       2.960       3.461 F_ACTIVE0                                             1.860       1.721       2.010       * For PROC DMREG (use node [Regression] in SAS EM), get:                                       Analysis of Maximum Likelihood Estimates                                                        Standard          Wald                  Standardized Parameter                         DF    Estimate       Error    Chi-Square    Pr > ChiSq        Estimate    Exp(Est)   Intercept                           1     -3.5620      0.1009       1247.02        <.0001                       0.028 F_ACTIVE0       0                   1     -0.3103      0.0198        246.48        <.0001                       0.733 F_REV           0                   1     -0.5817      0.0199        851.36        <.0001                       0.559 F_HV            0                   1      0.2257      0.0223        102.19        <.0001                       1.253 (The results for other 15 parameters almost match those from PROC LOGISTIC.)                          Odds Ratio Estimates                                                           Point Effect                                               Estimate F_ACTIVE0       0 vs 1                              0.538 F_REV           0 vs 1                              0.312 F_HV            0 vs 1                              1.570     Other 15 parameters almost have the same results of output statistics for both PROC LOGISTIC and PROC DMREG. However, the above 4 parameters (including Intercept) have different results.     One more thing, using PROC DMREG, why the statistics of 3 parameters for Exp (Est) are not equal to those for Point Estimate? (But other 15 parameters are equal.) ... View more

Help: why there are different results (Logistic Regression) from PROC ...

by Calcite | Level 5 in SAS Data Science
‎11-24-2017 10:57 AM
‎11-24-2017 10:57 AM
Hi,   I am using SAS Enterprise Miner and SAS Enterprise Guide to perform Logistic Regression upon the same dataset.   However, I get different set of output statistics from PROC DMREG (for Data Mining needs) in SAS EM compared to PROC LOGISTIC in SAS EG.   Which set of results is better? Is there any reference I can use to identify better results?     ... View more

When use PROC MERGE in SAS EG to merge 2 tables from Netezza, no need ...

by Calcite | Level 5 in SAS Enterprise Guide
‎11-17-2017 10:28 AM
‎11-17-2017 10:28 AM
Hi,   I am using SAS EG 4.3.   "For a merge with a BY statement, your data needs to be sorted, either by using PROC SORT or having data that is in sorted order already, or indexed using the BY variables."   But, when I use PROC MERGE to merge 2 tables from Netezza, I can skip over the step of PROC SORT. The output is correct and sorted. Why? Are the tables from Netezza pre-defined by something?     For example, (Netez is a libref for a Netezza database. 1st, create 2 tables in Netezza; 2nd, use PROC MERGE)   Data Netez.A1; Input ID Mth Name$ Height; cards; 7 5 D 2 1 1 A 1 3 2 B 2 9 5 E 2 5 3 C 2 ; run;   Data Netez.A2; Input ID Mth Name$ Weight; cards; 5 3 C 4 2 1 A 2 7 5 D 5 4 2 B 3 ; run;   data dummy; merge Netez.A1 (in=x) Netez.A2 (in=y); by id; if x; run; ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎02-16-2018 04:02 PM
Likes given to
View All
Quick links