Hi, there:     I want to change "Squarefilled" marker to "Square with horizontal bar" in the code & the picture below. data test; input x y lower upper @@; datalines; 1 10 8 11 2 11 10 14 3 13 10 14 4 13 11 15 5 14 12 16 ; run; proc sgplot data=test; scatter x=x y=y / yerrorlower=lower yerrorupper=upper markerattrs=(symbol=squarefilled) errorbarattrs=(color=blue pattern=solid); keylegend / noborder; run;   I want to change this like the following picture.   To use LEGENDITEM Statement in SGPLOT seems to be an answer for my question, but this statement applies to SAS9.4M5 and to later releases. And my SAS is 9.4M3. Please help me. Thank you in advance. ... View more

Hi, All:     I want to know How to Insert Custom Quantile Line to SGPLOT Graph. Example code is following. data test; input x y @@; datalines; 10 10 4 5 2 5 2 4 8 4 9 6 7 6 5 2 1 1 3 2 4 7 6 7 8 9 11 8 6 4 ; run; proc univariate data=test noprint; var x; output out=pctl pctlpts=25 50 75 100 pctlpre=pctl; run; data test_1; set test; if _n_ eq 1 then set pctl; if x le pctl25 then pctlgrp=1; else if x le pctl50 then pctlgrp=2; else if x le pctl75 then pctlgrp=3; else if x le pctl100 then pctlgrp=4; run; proc sort data=test_1 out=test_2; by pctlgrp; run; ods output basicintervals=mci(where=(parameter eq "Mean")); proc univariate data=test_2 cibasic; var y; by pctlgrp; run; proc univariate data=test_2 noprint; var x; by pctlgrp; output out=med median=med; run; data mci_1; set mci(drop=varname parameter); run; data test_3; merge test_2(in=a) med(in=b) mci_1(in=c); by pctlgrp; if a; run; proc sgplot data=test_3; reg x=x y=y / clm nomarkers; scatter x=med y=estimate / yerrorlower=lowercl yerrorupper=uppercl; run; On this graph, I want horizontal line (0th (x=1), 25th (x=3), 50th (x=6), 75th (x=8), 100th (x=11) percentile with marker line) on axis Y's "0.0", like following picture's black line. I tried to insert this line by making graph from specific data, but markers "PLUS" were out of this line's end. Thank you for your kind cooperation.       ... View more

Hi, All:     I'm making linear regression graph (with xx%CI), and I have trouble changing xx%CI's display. I want xx%CI graph with dashed line (which expresses confidence limit) like Prediction Limit's display, but now, I have regression line graph with xx%CI's shaded area by submitting following code. data test; input x y @@; datalines; 10 10 4 5 2 5 2 4 8 4 9 6 7 6 5 2 1 1 3 2 4 7 6 7 8 9 11 8 6 4 ; run; proc sgplot data=test; reg x=x y=y / cli clm clmattrs=(clmlineattrs=(pattern=dash color=black)); keylegend / noborder location=outside position=top; run; I tried by using CLMLINEATTRS= option, but I can't understand what changed. I want xx%CI's display which is expressed by dashed line & no shaded area, and also want to change xx%CI's label 's symbol (from shaded square to dash line).   By using Graphics Designer & following Generated Code, I accomplished my objective. But I want to accomplish it by PROC SGPLOT, not by PROC TEMPLATE&SGRENDER. ods output outputstatistics=pred(keep=observation predictedvalue lowerclmean upperclmean rename=(predictedvalue=pred lowerclmean=lower upperclmean=upper)); proc reg data=test; model y=x / r clm alpha=.1; run; quit; data test_1(drop=observation); merge test pred; run; proc template; define statgraph sgdesign; dynamic _HEIGHT _UPPER _LOWER _HEIGHT2 _PREDICT _HEIGHT3 _WEIGHT; begingraph; layout lattice / rowdatarange=data columndatarange=data rowgutter=10 columngutter=10; layout overlay; bandplot x=_HEIGHT limitupper=_UPPER limitlower=_LOWER / name='band' display=(OUTLINE) connectorder=axis extend=true outlineattrs=(color=CX000000 pattern=MEDIUMDASH ); seriesplot x=_HEIGHT2 y=_PREDICT / name='series' connectorder=xaxis; scatterplot x=_HEIGHT3 y=_WEIGHT / name='scatter'; discretelegend 'scatter' 'series' 'band' / opaque=false border=false halign=center valign=top displayclipped=true down=1 order=columnmajor location=inside; endlayout; endlayout; endgraph; end; run; proc sgrender data=test_1 template=sgdesign; dynamic _HEIGHT="x" _UPPER="upper" _LOWER="lower" _HEIGHT2="x" _PREDICT="pred" _HEIGHT3="x" _WEIGHT="y"; run;   Thank you for your kind cooperation. ... View more

Hi, All:     I'm trying to investigate clinical trial data by Linear Mixed-Effects Modeling and I want to make Predicted Mean Graph with 90%CI, which is like attached picture (Plots are not needed).  I started hunting and got following ideas, but I haven't reached clear one yet, and they are like difficult ones to understand. If you have easier ideas, advice, reference documents and so on, please let me know. Thank you for your kind cooperation. Ideas I reached are following..... No. 1: Use OUTP= option in MODEL Statement of PROC MIXED -> Use SGPLOT (I couldn't get wanted one.) No. 2: PROC MIXED or PROC GLIMMIX -> PROC PLM (difficult one? less information) /* For Example, I want to make graph by using information in Output "Solution for Fixed Effects" */ data hospital; input hospital physician x y @@; datalines; 1 1 19.4 32.4 1 1 19.3 28.3 1 1 19.2 31.9 1 2 18.8 18.9 1 2 18.4 20.3 1 2 18.2 21.3 1 3 19.7 24.8 1 3 20.4 38.6 1 3 20.2 27.4 1 4 19.3 36.1 1 4 21.7 32.4 1 4 20.8 30.7 2 1 21.5 31.3 2 1 18.7 28.5 2 1 19.9 20.1 2 2 18.9 31.8 2 2 18.7 25.1 2 2 19.5 32.9 2 3 18.7 18.3 2 3 19.4 27.4 2 3 19.5 26.2 2 4 20.9 28.6 2 4 21.1 29.9 2 4 20.0 16.4 ; run; proc mixed data=hospital; class hospital physician; model y=x / ddfm=kr solution; random int / subject=physician(hospital); run; Thank you. ... View more

Hi, All:   I want to make a figure, which expresses linear regression analysis. By submitting the following code, we can see one legend in the output, which includes "regression" & "95% confidence interval". data test; input x y @@; datalines; 10 10 4 5 2 5 2 4 8 4 9 6 7 6 5 2 1 1 3 2 4 7 6 7 8 9 11 8 6 4 ; run; proc sgplot data=test; reg x=x y=y / clm; keylegend / noborder location=outside position=top; run; By using LEGENDLABEL= option, I succeeded to change one value "regression" to "Mean predicted". But I want to change both legend values ("regression" & "95% confidence interval"). Please give me some advice.   Thank you for your kind cooperation. ... View more