topic Re: Using estimate statement in proc surveyreg in Statistical Procedures

Using estimate statement in proc surveyreg

kukie — Mon, 19 Oct 2020 20:19:20 GMT

Re: Using estimate statement in proc surveyreg

kukie — Mon, 19 Oct 2020 20:25:19 GMT

Hi, I am trying to find the estimates related to the effects of gender on the relationship between two variables.

I used an estimate statement in the code as follows....

proc surveyreg data=nhanes4 order=formatted;

cluster SDMVPSU;

strata SDMVSTRA;

weight WTSH2YR;

class vaper riagendr age race income cigs bmi;

model loglead= vaper|riagendr ridageyr bmxbmI race income cigs/

solution clparm;

format vaper vaperf. riagendr genderf. race racef. income

incomef. cigs cigsf. ;

estimate 'comparing males and females' riagendr -1 1/exp;

title "Model with Gender Interaction Term";

run;

somehow, I am not able to get the estimates for males and the estimates for females. Rather, it gives me a single estimate with male as the reference. Is there another way I can code this to give me two separate estimates, one each for male and females?

thanks

Re: Using estimate statement in proc surveyreg

SAS_Rob — Tue, 20 Oct 2020 12:49:21 GMT

Since gender is involved in an interaction you really can't look at a single parameter estimate to get the estimates for each of the two groups. In this case it is probably easiest to use the LSMEANS statement.

lsmeans riagendr;

Re: Using estimate statement in proc surveyreg

SteveDenham — Tue, 20 Oct 2020 13:03:03 GMT

I think what you are calling estimates are estimates of the parameters, which under the non-full rank parameterization used, sets the last category to zero. Using the LSMEANS statement combines the necessary elements to calculate the expected population mean for each level of a categorical variable.

This is a case where using the LSMEANS and LSMESTIMATE statement is an advantage; Try adding this to your code:

lsmeans vaper|riagender/e;
lsmestimate riagender "Males vs Females, averaged over vaper status' -1 1/exp;
lsmestimate vaper*riagender 'Males vs Females, vaper status=no'  -1 1 0 0,
                                                'Males vs Females, vaper status=yes' 0 0 -1 1/exp;

You may need to rearrange the coefficients in the second LSMESTIMATE statement to reflect the parameterization given in the LSMEANS statement by the /e option.

SteveDenham

Re: Using estimate statement in proc surveyreg

kukie — Fri, 23 Oct 2020 17:10:12 GMT

Hi Steve,

Thanks for this. The challenge I have now is that I have three levels of vapers - never vapers, former vapers and current vapers and I am struggling with the parameterization. I am unsure how to account for the three levels.

Thank you

Re: Using estimate statement in proc surveyreg

SteveDenham — Fri, 23 Oct 2020 17:23:50 GMT

That should not present a problem - once again it is why LSMESTIMATE statements are so useful.

lsmeans vaper|riagender/e;
lsmestimate riagender "Males vs Females, averaged over vaper status' -1 1/exp;
lsmestimate vaper*riagender 'Males vs Females, vaper status=never'  -1 1  0 0  0 0,
                            'Males vs Females, vaper status=former' 0 0  -1 1  0 0,
                            'Males vs Females, vaper status=current' 0 0  0 0  -1 1/exp;

This now gets trickier. I just stuck these in, and the correct version depends on the ordering of vaper levels. Since there is nothing in the CLASS statement and I assume that the levels are in fact 'Current', 'Former', and 'Never', the lsmeans will sort in alphabetical order. That means the code should look like:

lsmeans vaper|riagender/e;
lsmestimate riagender "Males vs Females, averaged over vaper status' -1 1/exp;
lsmestimate vaper*riagender 'Males vs Females, vaper status=current'  -1 1  0 0  0 0,
                            'Males vs Females, vaper status=former' 0 0  -1 1  0 0,
                            'Males vs Females, vaper status=never' 0 0  0 0  -1 1/exp;

If there is another coding for the vaper levels, it can be handled once you know the order that things appear in the LSMEANS statement.

SteveDenham.

Re: Using estimate statement in proc surveyreg

kukie — Mon, 26 Oct 2020 22:50:36 GMT

Thank you Steve,