topic Re: Symbol Adjustment in Proc Reg plots in ODS and Base Reporting

Symbol Adjustment in Proc Reg plots

DeannaGregory — Mon, 04 Jun 2018 23:43:44 GMT

Greetings SAS Community!

I think I know the answer to this question, but I am sincerely hoping I am wrong. I am creating plots for my ridge traces and variance inflation factors using Proc Reg. However, the output needs to be in grayscale for printing. Therefore, I need to change the symbols on the lines associated with the different variables. Is this possible? If so, how do I do it?

PS: I am using SAS 9.4

I am using a SAS sample dataset. Here is my code:

/* Example of Multicollinearity Findings */
libname health "C:\Program Files\SASHome\SASEnterpriseGuide\7.1\Sample\Data";

data health;
set health.lipid;
run;

/* Assess Pairwise Correlations of Continuous Variables */
proc corr data=health;
var age weight cholesterol triglycerides hdl ldl height;
title 'Health Predictors - Examination of Correlation Matrix';
run;

proc reg data=health;
model cholesterolloss = age weight cholesterol triglycerides hdl ldl height / vif tol collin;
title 'Health Predictors - Multicollinearity Investigation of VIF and Tol';
run;

/* Ridge Regression Example */
proc reg data=health outvif plots(only)=ridge(unpack VIFaxis=log)
outest=rrhealth ridge=0 to 0.10 by .002;
model cholesterolloss = age weight cholesterol triglycerides hdl ldl height;
plot / ridgeplot nomodel nostat;
title 'Health - Ridge Regression Calculation';
run;

Any help/guidance is sincerely appreciated!!

Re: Symbol Adjustment in Proc Reg plots

Reeza — Mon, 04 Jun 2018 23:59:28 GMT

Are you using the HTML output? If so, changing your style will modify the colours.

Try

ODS HTML style = journal;

You can put that at the top of your program and check the results. There’s also JOURNAL2/3 which are slightly different.

@DeannaGregory wrote:

Greetings SAS Community!

I think I know the answer to this question, but I am sincerely hoping I am wrong. I am creating plots for my ridge traces and variance inflation factors using Proc Reg. However, the output needs to be in grayscale for printing. Therefore, I need to change the symbols on the lines associated with the different variables. Is this possible? If so, how do I do it?

PS: I am using SAS 9.4

I am using a SAS sample dataset. Here is my code:

/* Example of Multicollinearity Findings */
libname health "C:\Program Files\SASHome\SASEnterpriseGuide\7.1\Sample\Data";

data health;
set health.lipid;
run;

/* Assess Pairwise Correlations of Continuous Variables */
proc corr data=health;
var age weight cholesterol triglycerides hdl ldl height;
title 'Health Predictors - Examination of Correlation Matrix';
run;

proc reg data=health;
model cholesterolloss = age weight cholesterol triglycerides hdl ldl height / vif tol collin;
title 'Health Predictors - Multicollinearity Investigation of VIF and Tol';
run;

/* Ridge Regression Example */
proc reg data=health outvif plots(only)=ridge(unpack VIFaxis=log)
outest=rrhealth ridge=0 to 0.10 by .002;
model cholesterolloss = age weight cholesterol triglycerides hdl ldl height;
plot / ridgeplot nomodel nostat;
title 'Health - Ridge Regression Calculation';
run;

Any help/guidance is sincerely appreciated!!

Re: Symbol Adjustment in Proc Reg plots

DeannaGregory — Tue, 05 Jun 2018 01:06:47 GMT

*forehead smack* YES, beautiful. That is exactly what I needed. I feel silly. Thank you!!!