proc arima data= SASHELP.HEART;

[knjohns0]

Can anyone please help me with the understanding  and the use  of proc arima data= SASHELP.HEART;. What is it used/purpose for and its feature. 

Thanks, 

Kiera 

[Cynthia_sas]

Hi:

  It would be more useful if you could explain what it is you want/need to know.

  If you don't understand the purpose of PROC ARIMA or how to use it, here's the documentation site: https://support.sas.com/documentation/onlinedoc/ets/151/arima.pdf and here's a clip from the Overview in that document:

  If you need more information about SASHELP.HEART, look it up here: https://support.sas.com/documentation/tools/sashelpug.pdf .

Cynthia

[knjohns0]

I am Analyzing Framingham Heart Data, the correlation between smoking and cholesterol data. 

[Reeza]

Arima is for time series analysis, first decide on your method of analysis, ie linear regression, logistic regression and then we can help you with the coding part. Coding is relatively easy, figuring out what to code is hard.

[knjohns0]

I am thinking Linear, since I am looking for a cause and effect. So, my study is based on the effects smoking has on cholesterol if any.

[Reeza · Accepted solution]

---

@knjohns0 wrote:

I am thinking Linear, since I am looking for a cause and effect. So, my study is based on the effects smoking has on cholesterol if any.

---

This should get you started and finished. It's more than you need, but a good example of looking at the data and doing an exploratory data analysis. There are missing data - I haven't attempted to account for that in any manner, just exploratory. Good Luck. 

ods graphics on;
title 'Exploratory Data Analysis';
title2 'Check Distributions';
proc freq data=sashelp.heart;
table chol_status smoking_status / missing;
table smoking cholesterol / missing;
run;

title2 'Ensure missing is consistent across variables';;

proc freq data =sashelp.heart;
table cholesterol*chol_status / missing;
table smoking*smoking_status / missing;
run;


title2 'Check distributions';
proc univariate data=sashelp.heart plots;
histogram smoking cholesterol;
var smoking cholesterol;
run;

title2 'Check distributions by status levels';
proc univariate data=sashelp.heart plots;
class chol_status;
var smoking ;
run;

title2 'Check distributions by status levels';
proc univariate data=sashelp.heart plots;
class smoking_status;
var cholesterol ;
run;


title2 'ANOVA comparisons and statistics';
proc anova data=sashelp.heart plots(maxpoints=10000);
class smoking_status; 
model cholesterol = smoking_status;
run;

title2 'Chi Square tests - status only';
proc freq data=sashelp.heart ;
table chol_status * smoking_status/ missing chisq;
run;

title2 'Regression modelling with continuous variables';
proc reg data=sashelp.heart plots(maxpoints=10000);
model cholesterol = smoking;
run;quit;

[knjohns0]

Thanks, for your help.