Hi SAS friends,

Following is a program that I adapted from the SAS STAT documentation for PROC LOESS. What's perplexing is that my output has a "connect the dots" appearance, rather than a single smoothed regression that centers in the data. Anyone know why this would occur?

many thanks!

Michael Murff, SLC, Utah

**eliminates duplicate matches for US file; *preparation for proc loess; **to address multiple lines in graphs; **LOESS or local regression is used when data contains outliers and robustness is desired; **beware of overfitting by choosing an smoothing parameter that is too small; **SAS defalut for alpha=.5 although it will optimially select alpaha too; **requires the data set "usjob" to be present in the work libarary; **run WW_merge.sas first in order to create usjob; **COP has used "Incumbent Weighted Salary" in the past; **A strong preference exists to continue use of this depvar; **"Incumbent Weighted Salary" = wsal;

**eliminate duplicates;

data usjob temp1; set wwjob(keep=id sheet_code wsal hay); where sheet_code='US'; by id; if last.id then output usjob; else output temp1; run;

proc freq data=usjob; tables id; run;

**check for duplicate ids; proc freq data=usjob; tables id; run;

**basic local regression;

proc loess data=usjob; model wsal=hay/details(OutputStatistics); run;

**introduce ods statement to generate output dataset;

proc loess data=usjob; model wsal=hay; ods output OutputStatistics=Results; run;

**check predictions and output;

title1 ’First 5 Observations of the Results Data Set’; proc print data=Results(obs=5); id obs; run;

symbol1 color=black value=dot; symbol2 color=black interpol=join value=none;

/* macro used in subsequent examples */

%let opts=vaxis=axis1 hm=3 vm=3 overlay; axis1 label=(angle=90 rotate=0);

proc gplot data=Results; title1 "Wage Data with Default LOESS Fit"; plot DepVar*hay Pred*hay/ &opts; run;

proc loess data=usjob; model wsal=hay/smooth=0.1 0.2 0.3 0.4 residual; ods output OutputStatistics=Results; run;

proc print data=Results(obs=5); id obs; run;

goptions nodisplay; proc gplot data=Results; by SmoothingParameter; plot DepVar*hay=1 Pred*hay/ &opts name= "fit"; run; quit;

goptions display; proc greplay nofs tc=sashelp.templt template=l2r2; igout gseg; treplay 1:fit 2:fit2 3:fit1 4:fit3; run; quit;