Adjusted R^2 (adjusted R square)

[Reeza]

And my original answer today was

@umarkhan  You didn't add the r=stw_residual to the PROC NLIN program in the output statement.
The PROC summary should point to this data set instead of the smallp dataset in Paige Miller code.

[afgdurrani0]

Full code:

data large;

input sow stw time;

datalines;

4 3.366 3

4 3.052 6

4 2.666 9

4 2.755 12

4 2.203 15

4 1.886 18

3.982 1.525 21

3.882 1.547 24

3.393 0.81 27

3.586 0.76 30

2.295 0.27 33

3.475 0.67 36

;

proc nlin data=large method=marquardt;

parms B=0.5 R=0.1;

delta=0.000001;

s=sow**(1-B)-R*(1-B)*time;

if s>0 then stx=s**(1/(1-B));

else stx=0;

model stw=stx;

sb=(sow**(1-B+delta)-R*(1-B+delta)*time);

if sb>0 then sdb=(stx-sb**(1/(1-B+delta)))/delta;

else sdb=0;

der.B=sdb;

sr=(sow**(1-B)-(R-delta)*(1-B)*time);

if sr>0 then sdr=(stx-sr**(1/(1-B)))/delta;

else sdr=0;

der.R=sdr;

output out=largep p=pstw r=stw_residual;

run;

proc summary data=largep;

     var stw_residual stw;

     output out=stats css(stw)=sstot uss(stw_residual)=ssres;

run;

data final;

     set stats;

     rsquared=1-(ssres/sstot);

run;

[afgdurrani0]

Did you read my full code? There no result about R^2 still.

[Reeza]

UmarKhan wrote:

Did you read my full code? There no result about R^2 still.

Yes. My original answer was wrong, but you didn't implement it anyways.

Did you open the final dataset?

proc print data=final;run;

[afgdurrani0]

It finally works Many thanks to Paige Miller and Reeza. the generated R^2 is Coefficient of determination or adjusted R^2?

[Reeza]

should get the correct answer not me, if you're able to change that.

[afgdurrani0]

Dear Reeza

The R^2 that is produced with this proc modification is adjusted R^2 or coefficient of determination?  Please answer.

[afgdurrani0]

Dear Paige Miller

The R^2 that is produced with this proc is adjusted R^2 or coefficient of determination?  Please answer

[Rick_SAS]

Paige Miller sent you a link to a Wikipedia article. If you read the article, you can learn about the R-squared statistic and see that the formula is 1-(SSres/SStot).  That same article contains a section about the adjusted R-squared statistic, in the event that you want to compute that statistic as well.

[afgdurrani0]

Dear Rick

I try but couldn't fit that formula in it.

Can you please modify it more with this for adjusted r^2. I need to write adjusted r^2 in my research paper.

I took the formula for adjusted R²  and can  be calculated as 1-[(1-R²)(n-1)(n-k-1)^-1]

where n is the number of data points and k is the number of regressors (parameters in your cases)

- and you already now are able to calculate R^2.

But I don't have any good knowledge of these programming, that's I need yous help in modifying the program further to produce adjusted r2.

Many thanks for the response. 

[Rick_SAS]

I feel confident that you can complete this step on your own if you give it a try. Time for you to try to learn some SAS programming! This is an easy computation, so it is a good opportunity for you to practice.

In the DATA step, right after you compute the rsquared variable, define a new variable like this:

adjrsquared = /* put formula here */;

In your formula, use the rsquared variable from the previous line, and substitute values for n and k from your problem. For example, if you have n=100 nonmissing abservations and k=2 parameters in your model, you would use 100 and 2 in the formula.

You should also know that multiplication is performed using the '*' symbol and division by using the '/' symbol. I suggest you use division instead of raising the last quantity to the -1 power.  And be sure to end each SAS statement with a semicolon.  Good luck!

[afgdurrani0]

Si, no result just errors !

proc summary data=expp;

     var stw_residual stw;

     output out=stats css(stw)=sstot uss(stw_residual)=ssres;

run;

data expp;

     set stats;

     rsquared=1-(ssres/sstot);

run;

proc print data=expp;

run;

data expp;

adjrsquared = (1-((1-rsquared)(16-2-1)/1));

run;

[Reeza]

Multiplication is denoted by *.

[afgdurrani0]

No error but no result

data expp;

adjrsquared = (1-(1-rsquared*16-1**16-2-1)/1);

run;

proc print data=expp;

run;

[Reeza]

You didn't follow the instructions from Rick.

In the DATA step, right after you compute the rsquared variable, define a new variable like this:
adjrsquared = /* put formula here */;

Add your line In the DATA FINAL step from the working code, not a new data step.

Also, I think your formula is incorrect.

EDIT: I would calculate it by hand and then tweak the program until I generated the correct results.

Additionally, if you go to the SAS Analytics U portion of this site there are FREE training courses on programming in SAS as well as YouTube videos.