Solved
Contributor
Posts: 70

# Skip data

Hello

I am asking for help which I want to explain in the following example:

I have data of 100 firms for the period 2000-2008. the data includes profit and No of directors for every firm.

I want to estimate a model that takes the profit data of years 2001,2003,2005,2007 directors data of the years 2000,2002,2004,2006.

Which functions in SAS I have to read in order to proceed with it

Thanks a lot!!!!

Accepted Solutions
Solution
‎06-16-2016 01:33 PM
Posts: 5,543

## Re: Skip data

Here is an example where I model Weight as a function of Age, only for odd ages and use the model to estimate the weight of students of even ages.

``````data regClass;
set sashelp.class;
/* Decide which ages to include in the model */
includedAge = mod(age,2);
/* Copy only included weights, other will be set to missing */
if includedAge then regWeight = weight;
run;

/* Estimate the model, output predictions for all observations */
proc reg data=regClass plots=none;
model regWeight = age;
output out=predClass p=predWeight;
run;

proc sort data=predClass; by age; run;

proc sgplot data=predClass;
scatter x=age y=weight / group=includedAge;
series x=age y=predWeight;
run;
``````

of course, proc reg is only one of many procedures available to build predictive models, but most procs work about the same way.

PG

All Replies
Super User
Posts: 9,606

## Re: Skip data

Without knowing what the data looks like its hard to say, is there a variable for year, if so then just where clause that:
where year in (2001,2003,2005,2007);

For example.  To get good answers, it a good idea to post test data, in the form of a datastep, and what the output should be.  Posting the relevant code in this instance would also illustrate it.

Solution
‎06-16-2016 01:33 PM
Posts: 5,543

## Re: Skip data

Here is an example where I model Weight as a function of Age, only for odd ages and use the model to estimate the weight of students of even ages.

``````data regClass;
set sashelp.class;
/* Decide which ages to include in the model */
includedAge = mod(age,2);
/* Copy only included weights, other will be set to missing */
if includedAge then regWeight = weight;
run;

/* Estimate the model, output predictions for all observations */
proc reg data=regClass plots=none;
model regWeight = age;
output out=predClass p=predWeight;
run;

proc sort data=predClass; by age; run;

proc sgplot data=predClass;
scatter x=age y=weight / group=includedAge;
series x=age y=predWeight;
run;
``````

of course, proc reg is only one of many procedures available to build predictive models, but most procs work about the same way.

PG
🔒 This topic is solved and locked.

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

Discussion stats
• 2 replies
• 320 views
• 0 likes
• 3 in conversation