Programming the statistical procedures from SAS

Statistical significance of time trends

Accepted Solution Solved
Reply
Occasional Contributor
Posts: 13
Accepted Solution

Statistical significance of time trends

Hi SAS community,

I was wondering if you guys can help me out with the correct approach in determining whether or not a time trend is statistically significant over time.

Lets say in year 2000, we had 100 cases with over population of 50,000 = 200/100,000 persons/year.

in year 2005, we had 75 cases with populatino of 60,000 = 125/100,000 persons/year.

in year 2010, we had 50 cases over a population of 75,000 = 66.67/100,000 persons/year.

How do I get the p value and determine whether it's statistically different?

Thanks in advance for all your help!


Accepted Solutions
Solution
‎04-13-2013 02:58 PM
Regular Contributor
Posts: 152

Re: Statistical significance of time trends

Another possibility is Poisson regression using PROC GENMOD,

   data test(keep=year case log_n);

       infile datalines;

       input year case n;

       log_n=log(n);

       output test;
   datalines;

   2000 100  50000

   2005   75   60000

   2010   50   75000

   ;

   run;

   proc genmod data=test;

       model case=year / dist=poisson link=log offset=log_n;

   run;

View solution in original post


All Replies
Respected Advisor
Posts: 4,606

Re: Statistical significance of time trends

You can test for a linear time trend (Cochran-Armitage Test for Trend) as follows :

data test(keep=year case n);
input year cases population;
case = "YES";
n = cases; output;
case = "NO";
n = population-cases; output;
datalines;
2000 100 50000
2005 75 60000
2010 50 75000
;

proc freq data=test;
weight n;
tables year*case / trend;
run;

PG

PG
Respected Advisor
Posts: 4,606

Re: Statistical significance of time trends

Note : The numbers used for testing should reflect actual observations, not population estimates. - PG

PG
Solution
‎04-13-2013 02:58 PM
Regular Contributor
Posts: 152

Re: Statistical significance of time trends

Another possibility is Poisson regression using PROC GENMOD,

   data test(keep=year case log_n);

       infile datalines;

       input year case n;

       log_n=log(n);

       output test;
   datalines;

   2000 100  50000

   2005   75   60000

   2010   50   75000

   ;

   run;

   proc genmod data=test;

       model case=year / dist=poisson link=log offset=log_n;

   run;

☑ This topic is SOLVED.

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

Discussion stats
  • 3 replies
  • 1165 views
  • 3 likes
  • 3 in conversation