Help using Base SAS procedures

multiple linear regression

Accepted Solution Solved
Reply
Contributor
Posts: 65
Accepted Solution

multiple linear regression

[ Edited ]

I have this program:

 

 

proc reg data=TRNSTRANSPOSEDQUERY_FOR_TRNSTRAN;
model LOG_ALLWD = ABDOM_HERNIA ADV_DIABETES CHF_CARDIOMYO_VALVDIS CHRON_RENAL_FAIL DEP_BIPOLAR_PARA DIAG_W_ADD DVT_PE GASTRDUO_ULCER GAST_DUODENITIS GI_BLEED HEP_CIRR_OTR_LIVER IBS INTEST_OBSTR MILD_SLEEP_APNEA NUTR_OTR_ANEMIA RESP_FAILURE RHEU_ARTH SCREEN_W_ADD SHOCK_SYNC STROKE_PARAL TOBACCO
/ SELECTION=BACKWARD
SLS=0.15
ALPHA=0.1 ;
title "Backward Selection Model";
run; quit;

 

and got this report: I want to save this report to a sas data set in the same format. I used the option outset=estimate and did not get what I want. Do you have a way to do this? Thanks!

Variable Parameter
Estimate Standard
Error Type II SS F Value Pr > F
Intercept 7.05997 0.00517 151738 1863176 <.0001
ABDOM_HERNIA 0.11768 0.04140 0.65785 8.08 0.0045
ADV_DIABETES 0.03687 0.01479 0.50618 6.22 0.0127
CHF_CARDIOMYO_VALVDIS 0.09931 0.02582 1.20509 14.80 0.0001
CHRON_RENAL_FAIL 0.11154 0.04609 0.47694 5.86 0.0155
DEP_BIPOLAR_PARA 0.09403 0.02506 1.14656 14.08 0.0002
DIAG_W_ADD 0.30107 0.00906 89.88892 1103.74 <.0001
DVT_PE 0.05880 0.03807 0.19423 2.38 0.1225
GASTRDUO_ULCER 0.04103 0.03024 0.14988 1.84 0.1750
GAST_DUODENITIS 0.26914 0.01104 48.38799 594.15 <.0001
GI_BLEED 0.08496 0.01131 4.59823 56.46 <.0001
HEP_CIRR_OTR_LIVER 0.14005 0.02110 3.58960 44.08 <.0001
IBS 0.11712 0.01032 10.49936 128.92 <.0001
INTEST_OBSTR 0.08401 0.01877 1.63125 20.03 <.0001
MILD_SLEEP_APNEA 0.04275 0.01265 0.92967 11.42 0.0007
NUTR_OTR_ANEMIA 0.14219 0.01517 7.15148 87.81 <.0001
RESP_FAILURE 0.14960 0.02995 2.03182 24.95 <.0001
RHEU_ARTH 0.02965 0.01502 0.31725 3.90 0.0485
SCREEN_W_ADD 0.24392 0.00721 93.13549 1143.60 <.0001
SHOCK_SYNC 0.20405 0.04136 1.98179 24.33 <.0001
STROKE_PARAL 0.07213 0.04339 0.22507 2.76 0.0965
TOBACCO 0.10960 0.01568 3.97830 48.85 <.0001

Accepted Solutions
Solution
‎03-30-2017 10:00 AM
Super User
Posts: 19,772

Re: multiple linear regression

Posted in reply to zhuxiaoyan1

Have you tried the ODS table instead?

 

See instructions here on how to find the name and save it. My guess is it would be called ParameterEstimates.

 

http://blogs.sas.com/content/iml/2017/01/09/ods-output-any-statistic.html

View solution in original post


All Replies
Solution
‎03-30-2017 10:00 AM
Super User
Posts: 19,772

Re: multiple linear regression

Posted in reply to zhuxiaoyan1

Have you tried the ODS table instead?

 

See instructions here on how to find the name and save it. My guess is it would be called ParameterEstimates.

 

http://blogs.sas.com/content/iml/2017/01/09/ods-output-any-statistic.html

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 1 reply
  • 176 views
  • 1 like
  • 2 in conversation