- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello ,
How could i find dependency/importance/weightage between Dependent variables in SAS?
Example:
Var1 Var2 Result
1 2 5
3 2 7
in the above example weigtage of var2 is higher because the formala is result=var1+2var2
Thanks,
Mushy
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Your question is not clear, but if you are asking how to assess the relative importance of predictor (independent) variables in a model on a response (dependent) variable, then see this note.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
The first step for you question is summary statistic by exploring scatter plot.
proc sgscatter data=sashelp.heart(obs=100);
matrix height weight Diastolic Systolic/ellipse diagonal=(histogram);
run;
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
@Mushy wrote:
How could i find dependency/importance/weightage between Dependent variables in SAS?
Example:
Var1 Var2 Result
1 2 5
3 2 7
in the above example weigtage of var2 is higher because the formala is result=var1+2var2
I find this confusing sequence of statements. I am not sure what you want.
You start by asking about correlation between independent and dependent variables in the subject line, and then only talk about dependent variables "How could i find dependency/importance/weightage between Dependent variables in SAS?"
You state the fact "in the above example weigtage of var2 is higher because the formala is result=var1+2var2" but this isn't a question; what is the relevance of this?
Paige Miller
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Your question is not clear, but if you are asking how to assess the relative importance of predictor (independent) variables in a model on a response (dependent) variable, then see this note.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
@StatDave @Ksharp @PaigeMiller Thanks for the response. Your posts helped me to explore the statistics graph and used different procs for dependency check.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
I'm not sure what your may be thinking about asking with correlation but if I build a data set with your formula for some data and run a regression:
data examine; do x1= 1 to 25; do x2 = -5 to 5; result = x1 + 2*x2; output; end; end; run; proc reg data=examine; model result=x1 x2; run;
The coefficient for X2 is that "weight"
Parameter Estimates | |||||
---|---|---|---|---|---|
Variable | DF | Parameter Estimate |
Standard Error |
t Value | Pr > |t| |
Intercept | 1 | 0 | 0 | . | . |
x1 | 1 | 1.00000 | 0 | Infty | <.0001 |
x2 | 1 | 2.00000 | 0 | Infty | <.0001 |