Hi: Please do not bury your new posting in an 11 year old previous posting. You cannot use GREPLAY with PROC SGPLOT. PROC SGPLOT belongs to ODS GRAPHICS and GREPLAY belongs to SAS/GRAPH. ODS GRAPHICS images are NOT stored in a SAS/GRAPH catalog. I suggest you make a new posting on your topic and show the code you've tried or explain your task in detail. Cynthia ... View more

my model is simple, i regress individual stock liquidity on market liquidity for 500 stocks using following equation IL = a + b1 ML +e i regress it for every stock for every month over a period of 7 years, i extract the R2 from the above equations for every stock in every month, calculate the cross sectional average every month. hence i have a time series monthly R2 for a period of 7 years, which represent the liquidity commonality of the market. as R2 is bouded between 0 and 1, i perform a logit transformation. i understand that the values obtained from logit transformations are all negative because most of the values of orignal R2 are between .04 and .2 now i want to regress this using OLS , on the monthly market returns, abs + monthly market returns, and abs negative monthly market returns. monthly market returns have mostly the value between 0.1 and .2, but i get coefficients like 100 and 22 which make absolutely no economic sense. I dont know how to interpret the coefficients, or if i am doing something wrong while estimating the last equation. Thanks but i ge ... View more

A possibility is to extend 's idea of a shift to GLM even farther to use MIXED.  The following is untested, but might work: ods output corrb=corrb solutionf=solutionf; proc mixed; class stockName; model liq = stockName x z d / solution noint corrb; run; The corrb option should give the pairwise correlations you need (there may need to be some nesting to get stuff right, but for now try this).  The regression coefficients should be in the solutionf dataset. Steve Denham Message was edited by: Steve Denham ... View more

thanks you are a great help ... View more

hi ... the SAS code is in the TEN_THINGS_SAS.ZIP that's attcahed (along with the PDF and PNG files) to my posting just in case that's still a problem ... proc sort data=sashelp.class out=class; by height; run; ods results off; ods listing close; * use startpage=never to keep adding to a single page; ods pdf file='z:\ten_things.pdf' notoc startpage=never; options orientation=landscape; * title for all ten "plots"; goptions reset=all border ftext='calibri' htext=2.5; proc gslide; title 'TEN "DIFFERENT" THINGS FROM SAS/GRAPH GPLOT AND GSLIDE ON ONE PAGE' ls=2; run; quit; * resize graphics output to 2x3 inches; goptions reset=all border ftext='calibri' gunit=pct htext=3 hsize=2in vsize=3in; symbol1 v=dot i=join h=2 c=blue; axis1 label=(a=90); * use horigin and vorigin to place output on the page; * 5 charts from GPLOT; proc gplot data=class; title 'PLOT #1' ls=2; goptions horigin=0.4in vorigin=1.0in; plot weight*height / vaxis=axis1; run; title 'PLOT #2' ls=2; goptions horigin=2.45in; plot weight*height / vaxis=axis1; run; title 'PLOT #3' ls=2; goptions horigin=4.5in; plot weight*height / vaxis=axis1; run; title 'PLOT #4' ls=2; goptions horigin=6.6in; plot weight*height / vaxis=axis1; run; title 'PLOT #5' ls=2; goptions horigin=8.65in; plot weight*height / vaxis=axis1; run; quit; * 5 charts from GLSIDE; proc gslide; title 'GSLIDE #1' ls=2; goptions horigin=0.4in vorigin=4.25in; run; title 'GSLIDE #2' ls=2; goptions horigin=2.45in; run; title 'GSLIDE #3' ls=2; goptions horigin=4.5in; run; title 'GSLIDE #4' ls=2; goptions horigin=6.6in; run; title 'GSLIDE #5' ls=2; goptions horigin=8.65in; run; quit; ods pdf close; ods listing; ods results; ... View more

Try this approach : /* create random test data for 5 stocks with missing values */ data test(drop=s); array stock{5}; format date date9.; do date = '01JAN2010'd to '31DEC2011'd;       if weekday(date) in (2,3,4,5,6) then do;             do s = 1 to dim(stock);                   stock{s} = ifn(rand('UNIFORM') < 0.9, 100 + 10*rand('NORMAL'), .);                   end;             output;             end;        end; run;      /* List the stock variable names */ %let stocks=stock1-stock5;      proc transpose data=test out=temp0; by date; var &stocks.; run;      proc sql; create table temp1 as select intnx("WEEK", date, 0) as week format=date9., _name_, mean(COL1) as meanStock from temp0 group by calculated week, _name_ having count(COL1) >= 4; drop table temp0; quit;      proc transpose data=temp1 out=want(drop=_name_); by week; id _name_; var meanStock; run; PG ... View more