Wald Confidence Interval for Adjusted Odds Ratios in PROC SURVEYLOGISTIC

[sstoline]

Dear All:

I do need your help with how to get Wald Confidence Interval for Adjusted Odds Ratios in the proc surveylogistic.

The following SAS code(s) only give me the Odds Ratio Estimates table with Point estimate and the 95% Wald Confidence Interval.


proc surveylogistic data =have;
strata STRATUM;
weight WEIGHT;
class x1(param=ref ref = "1") x2(param=ref ref = "1") x3(param=ref ref = "1") x4(param=ref ref = "1") x5(param=ref ref = "1") x6(param=ref ref = "1") x7(param=ref ref = "1") x8(param=ref ref = "1") x9(param=ref ref = "1") x10(param=ref ref = "2");
model  Y (event='1') = x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15;
run;

Response variable Y has two categories (1,2)

some of the independent variables (x1-x10) are categorical variables some have 2 categories (1,2), others have three categories (1,2,3), and x11-x15 are continuous variables.

Any helps will be highly appreciated.

with many thanks

steve

[sstoline]

I do need your help with how to get Wald Confidence Interval for Adjusted Odds Ratios in the proc surveylogistic.

The following SAS code(s) only give me the Odds Ratio Estimates table with Point estimate and the 95% Wald Confidence Interval.


proc surveylogistic data =have;
strata STRATUM;
weight WEIGHT;
class x1(param=ref ref = "1") x2(param=ref ref = "1") x3(param=ref ref = "1") x4(param=ref ref = "1") x5(param=ref ref = "1") x6(param=ref ref = "1") x7(param=ref ref = "1") x8(param=ref ref = "1") x9(param=ref ref = "1") x10(param=ref ref = "2");
model  Y (event='1') = x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15;
run;

Response variable Y has two categories (1,2)

some of the independent variables (x1-x10) are categorical variables some have 2 categories (1,2), others have three categories (1,2,3), and x11-x15 are continuous variables.

Any helps will be highly appreciated.

with many thanks

steve

[skg74_mail_umkc_edu]

Hi Steve,

ODS trace on;

ods output    ParameterEstimates=a****User defined output dataset name;

ods output    oddsratios=b****User defined output dataset name;

proc surveylogistic data =have;
strata STRATUM;
weight WEIGHT;
class x1(param=ref ref = "1") x2(param=ref ref = "1") x3(param=ref ref = "1") x4(param=ref ref = "1") x5(param=ref ref = "1") x6(param=ref ref = "1") x7(param=ref ref = "1") x8(param=ref ref = "1") x9(param=ref ref = "1") x10(param=ref ref = "2");
model  Y (event='1') = x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15;
run;

ODS output off;

ODS trace off;

Check newly created datasets a, b. You will find Wald statistics.

Best Regards,

Sri.