topic Re: ANCOVA with 3 categorical variable-SAS studio in SAS Studio

ANCOVA with 3 categorical variable-SAS studio

khem — Tue, 04 Aug 2020 19:22:15 GMT

Hi,
I am trying to do ANCOVA using the SAS studio for three categories based on data. Please may i know how can i use 3 categorical variables at the same time? My collaborators used the same data and used the SAS enterprise guide (which i don't have) and used 3 variables together. using 1 variable is not helping me to reproduce similar results. I need to reproduce the same results for further analysis.

Here i have attached the data and results from SAS enterprise guide,

In the data zero (o) means "without diet", and one (1) means "with diet".

Please help me, Thanks,
Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Tue, 04 Aug 2020 19:31:42 GMT

We really would benefit from seeing the code used by your collaborators. Neverthess, this code should produce an ANCOVA with 3 categorical variables (which I have named CAT1 CAT2 and CAT3) and one covariate (which I have named X)

proc glm data=have;
    class cat1 cat2 cat3;
    model y = x cat1 cat2 cat3;
run;
quit;

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Tue, 04 Aug 2020 20:24:01 GMT

Hi @PaigeMiller Thanks for the quick response,

I ran again with your suggested code,

I could up to typeIII values and F values, this is great, Exact same like my collaborator,

I attached my results too (SAS_studio...pdf)

But i am not able to get the estimate and t and Pr>t values (please check the attached file [ancova_cassie...] for results from a collaborator) like my collaborators had,

Please can you help me on this too,

They used SAS enterprise guide, so i am not sure, if they will have codes for them,

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Tue, 04 Aug 2020 21:27:12 GMT

Generally, I do not download attachments. Please post the relevant information as a screen capture in your message.

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Tue, 04 Aug 2020 23:58:29 GMT

Hi @PaigeMiller ,

Okay,

Here i dropped the screen images for the results of my collaborator (first one with blue font) and mine (image with black&white font). My results are based on your previous suggestion. I got similar values for TypeIII and Pr>F like my collaborator, which is great. Your suggestion is in the perfect and right direction for my data. Please kindly suggest me how can I get similar intercept and Pr>t values too? for the reference, I also put the image of my datasheet.

Note: Here In the data zero (o) means "without diet", and one (1) means "with diet"

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Wed, 05 Aug 2020 00:43:44 GMT

I don't see any similarity. It doesn't even look like you have the same data or the same response variable.

You really need to confirm you have the same data and are using the same code.

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Wed, 05 Aug 2020 00:49:44 GMT

Hi @PaigeMiller

I am really sorry,

I uploaded the wrong file of my collaborator,

Here i attached the right file of my collaborator, along with my same results (after your suggestion, black&whilte figure) and data from my previous post,

Here I got similar values for TypeIII and Pr>F like my collaborator, now i need to get same intercept, estimates, Pr>t values.

Sorry again,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Wed, 05 Aug 2020 00:54:22 GMT

You don't show the intercept or estimates for the 2nd analysis.

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Wed, 05 Aug 2020 04:08:51 GMT

Hi @PaigeMiller ,

Second analysis i did, based on your first suggestion but when i ran your equation in SAS studio, I didn't get the intercept, estimates and Pr>t values.

do i need to add something into your previous equation?

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Wed, 05 Aug 2020 10:40:25 GMT

oh, I understand now, that was my omission. Use this MODEL statement with the SOLUTION option

model y = x cat1 cat2 cat3/solution;

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Wed, 05 Aug 2020 19:22:37 GMT

Hi @PaigeMiller

Thanks for the suggestion, it worked,

Here is the image

I got the same estimate and Pr>t values for all three diets,

Only one doubt, the estimate of intercept is not matching,

My collaborator had 1.26 (in previous post figure) but mine is 2.60. Does it matter? do i need to make any changes?

Thanks again,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Wed, 05 Aug 2020 19:31:23 GMT

That's very odd, and I don't have an explanation about how everything can be the same except the intercept.

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Wed, 05 Aug 2020 19:47:09 GMT

Oh,

Here i attached my codes, if it may help you,

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Thu, 06 Aug 2020 05:21:17 GMT

Hi @PaigeMiller ,

Sorry to bother you again,

is it possible to reproduce your model for ANCOVA in R?

If yes, can you suggest to me how? what should i add for solution (SAS studio) option in R?

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

PaigeMiller — Thu, 06 Aug 2020 10:40:52 GMT

You need to ask your question in an R forum and not in a SAS forum.

Re: ANCOVA with 3 categorical variable-SAS studio

Rick_SAS — Tue, 11 Aug 2020 13:59:01 GMT

The simplest explanation for why your Intercept estimate differs from your colleague is that you used different centering values when you standardized the covariate (CRP_A). When you called PROC STDIZE, you centered by the data mean and data stdDev. If your colleague centered by another value, he/she would get a different intercept estimate.

Here is an example that shows the same PROC GLM statements. The first analysis centers the data by subtracting the value 20. The second call centers the X variable by subtracting the sample mean. All estimates are the same except for the Intercept.

data class;
set sashelp.class;
ageCat = (Age<13);
heightCat = (Height<62);
call streaminit(1);
x = rand("Normal", 20, 1);
run;

title "Colleague Analysis";
proc stdize data=class out=class1 add=-20;
var x;
run;

proc glm data=class1 plots=none;
class Sex ageCat heightCat;
model weight = x Sex ageCat heightCat / solution;
ods select ParameterEstimates;
quit;

/*------------------*/
proc stdize data=class out=class2;
var x;
run;

title "My Analysis";
proc glm data=class2 plots=none;
class Sex ageCat heightCat;
model weight = x Sex ageCat heightCat / solution;
ods select ParameterEstimates;
quit;

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Fri, 14 Aug 2020 05:45:54 GMT

Hello @Rick_SAS ,

Thanks for the explanation,

I tried to understand your example and compare it with my code,

Looks like my code is similar to your second example (please correct me if i am wrong),

If I want to get the same intercept estimate like my colleague (they used SAS enterprise and i don't have their code), do i need to remove the method part?

AS you defined X= rand...., i cant do guess for CRP_A

x = rand("Normal", 20, 1);

or add = -20

I am sorry, but i don't know what change to i need to make? i am still a newbie to SAS.

Please guide me, what change is needed in my code?

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

Rick_SAS — Fri, 14 Aug 2020 10:16:28 GMT

> If I want to get the same intercept estimate like my colleague

> (they used SAS enterprise and i don't have their code),...

1. If you want to get the same estimate, the easiest solution is to contact your colleague. This will enable you to understand what is happening.

2. If the problem is caused by different centering of X (and we do not know for certain that it is), then both models will give the same predicted variables. Only the parameterization is different, but they represent the same predictive model. Thus there is really no need to get the same estimates.

3. If you can't do #1 and you aren't convinced by #2 and you insist on getting the same answer, then center CRP_A by using the value -1.681563224. That is, use

ADD = -1.681563224

on the PROC STDIZE statement.

Be aware that #3 is a hack. The best solution is #1.

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Fri, 14 Aug 2020 17:12:33 GMT

Hi @Rick_SAS ,

Thanks for the tips,

I added ADD = -1.681563224 but i couldn't get the same intercept estimate (please see the image). By the way, how do you calculate the value for ADD?

I asked my colleague for the details, how i can get the code,

Just for curiosity, does it possible to get code from SAS enterprise guide version?

Thanks,

Khem

Re: ANCOVA with 3 categorical variable-SAS studio

khem — Fri, 09 Oct 2020 04:55:49 GMT

Hi @Rick_SAS @PaigeMiller

I am doing ANCOVA with three categorical variable. But i am confused about how to represent the results in a graph/plot,

a box/scatter, i dont know how it works, this is my first time doing using SAS studio,

Please someone can guide me on this, i would appreciate the support,

Thanks,