Gee error

[bee2]

Hi,

I am running gee for longitudinal analysis and am encountering a prob:

ERROR: No valid observations due to invalid or missing values in the response, explanatory, offset, frequency, or weight variable.

The code i am running is :

Proc genmod data=dietdiv2 ;
class hhid  peduc(ref='1');
Model stuntingall3= peduc/type3 dist=poisson link=log;
repeated subject=hhid/type=ind;
estimate 'peducn' peduc 1 -1 1/exp;
run;

The exposure (peduc) is measured at visit 1 (baseline) and the outcome(stuntingall) at visit 3:

SAS Output

stuntingall3 Frequency Percent CumulativeFrequency CumulativePercent01Frequency Missing = 19042

2907	71.62	2907	71.62
1152	28.38	4059	100.00

peduc Frequency Percent CumulativeFrequency CumulativePercent12Frequency Missing = 18150

2383	48.13	2383	48.13
2568	51.87	4951	100.00

Any thoughts on how i can address the errors?

Thanks.

[Reeza]

That’s a lot of missing. Is that what you expect?

---

@bee2 wrote:

Hi,

I am running gee for longitudinal analysis and am encountering a prob:

ERROR: No valid observations due to invalid or missing values in the response, explanatory, offset, frequency, or weight variable.

The code i am running is :

Proc genmod data=dietdiv2 ;
class hhid  peduc(ref='1');
Model stuntingall3= peduc/type3 dist=poisson link=log;
repeated subject=hhid/type=ind;
estimate 'peducn' peduc 1 -1 1/exp;
run;

The exposure (peduc) is measured at visit 1 (baseline) and the outcome(stuntingall) at visit 3:

SAS Output

stuntingall3 Frequency Percent CumulativeFrequency CumulativePercent01Frequency Missing = 19042

2907	71.62	2907	71.62
1152	28.38	4059	100.00

peduc Frequency Percent CumulativeFrequency CumulativePercent12Frequency Missing = 18150

2383	48.13	2383	48.13
2568	51.87	4951	100.00

Any thoughts on how i can address the errors?

Thanks.

---

[bee2]

Hi,
I expect to have a lot of missings given that i have 7 follow-up points and the exposure is measured at timepoint 1(baseline) and the outcome at timepoint 3. The responses for both the exposure and outcome are missing at the other time points. Hope that clarifies.

[sld]

Does that mean that you have only two values for each HHID, namely stuntingall3 and peduc? If so and if you are using stuntingall3 as the response and peduc as a predictor, then you do not have a data structure with repeated measures and the REPEATED statement is not necessary and would generate errors.

Looking ahead, if peduc has only two levels, then the ESTIMATE statement is not correct because it specifies 3 parameters.