Hi All,   is that possible to run out a scatter plot within multiple groups, also with regression line, coefficient and pvalue in one figure? I want use macro or dynamic proc template since I have lots of variables. I searched a lot, but didn't get a good example. Anyone can provide me some idea or example??   Thanks so much. C ... View more

Hi All,   I wrote a dynamic template to draw scatter plots, but I want to add coefficient and p value into the figure. The coefficient and p value can be anywhere in the figure. Like what I did now, it's above the figure, but I insert the values manually one by one. Any idea to input then quickly? my goal is just show the values in the figure, I don't care the type, the location or some other things. I just don't want input manually. Below is my template.Code with bold are what I want to change.   proc template; /* Create the plot template. */ define statgraph testPattern; begingraph / designheight = 8in; dynamic var1 var2 var3; layout lattice / columns = 1 rows = 2 rowgutter = 20 columngutter = 20; /* */ cell; cellheader; layout gridded / border = true; entry "Toddler S = 0.518 p = <.001; Young S = 0.383 p = <.001" / textattrs = (weight = bold); entry "Child S = 0.428 p = <.001; Teen S = 0.383 p = <.001" / textattrs = (weight = bold); endlayout; endcellheader; layout overlay / yaxisopts = (label = var1 labelattrs = (weight = bold size = 8pt) tickvalueattrs = (weight = bold)) xaxisopts = (label = var3 labelattrs = (weight = bold) tickvalueattrs = (weight = bold)); scatterplot y=var1 x=var3 / group=group; regressionplot y=var1 x=var3 / group = group name="lines"; discretelegend "lines"; endlayout; endcell; cell; cellheader; layout gridded / border = true; entry "Toddler S = 0.517 p = <.001; Young S = 0.383 p = <.001" / textattrs = (weight = bold); entry "Toddler S = 0.377 p = <.001; Young S = 0.318 p = 0.002" / textattrs = (weight = bold); endlayout; endcellheader; layout overlay / yaxisopts = (label = var2 labelattrs = (weight = bold size = 8pt) tickvalueattrs = (weight = bold)) xaxisopts = (label = var3 labelattrs = (weight = bold) tickvalueattrs = (weight = bold));; scatterplot y=var2 x=var3 / group=group; regressionplot y=var2 x=var3 / group=group name="lines"; discretelegend "lines"; endlayout; endcell; endlayout; endgraph; end; run; /* Generate the plot. */ options orientation=landscape; ods rtf file="scattertest1.rtf"; proc sgrender data=combine template='testPattern'; dynamic var1 = 'P_SCD_SCORE_PH' var2 = 'P_SCD_SCORE_PI' var3 = 'P_SCD_SCORE_PMC'; run; ods rtf close; ... View more

Hi All,   Can I run a dynamic proc template of histogram but with different groups? I have a variable, it has three groups, I can run with different groups in boxplot, but how can I edit histogram part by groups? Below is the code I use and the output.   proc template; define statgraph mygraphs.dynamics; begingraph; layout lattice / columns = 1 rows = 2 columndatarange = union rowgutter = 2px; mvar SYSDATE9 SCALE; nmvar BINS; dynamic VAR VARLABEL; entrytitle 'Histogram of ' VAR; entrytitle 'with Normal Distribution'; columnaxes; columnaxis / label = 'var'; endcolumnaxes; layout overlay / xaxisopts=(label=VARLABEL) yaxisopts = (offsetmin = .03); histogram VAR / display = standard; class group; densityplot VAR / normal( ); fringeplot var; endlayout; boxplot y = var x = group/ orient = horizontal; endlayout; entryfootnote halign=right "Created: " SYSDATE9 / textattrs=GraphValueText; endgraph; end; run; %let scale=percent; proc sgrender data=combine_1 template='mygraphs.dynamics'; dynamic var='P_SCALE_SCORE_PHYSICAL_FUNCTI' varlabel='P_SCALE_SCORE_PHYSICAL_FUNCTI'; run;   thanks, C ... View more

Hi All,   Is that possible to add p-value (signed rank) into a histogram picture ?  Below is my code and I attached a picture:   proc univariate data=combine_1 normal; var var_1; class Group; histogram var_1 / midpoints = (0 to 100 by 5) nrow = 3 odstitle = "VAR" normal (color = red) kernel cfill = ltgray ctext = blue nohlabel vscale = percent; inset N = 'N' MEDIAN (8.2) MEAN (8.2) STD = 'Standard Deviation'(8.3) / position = nw ; ods select histogram; run;   Thanks, C   ... View more

Hi All,   I want to output a rtf file with multiple figures, but I just know how to stack multiple graphs vertically on the same page, is that possible to have 2 columns in one page in rtf? I use layout, but the log shows layout is not supported in RTF destination. Below is my code:   options nodate nonumber; ods _all_ close; ods rtf file = 'figures.rtf' startpage = no image_dpi = 300; ods layout gridded columns = 2; ods graphics / reset = all outputfmt = png height = 3in width = 4in; proc univariate data=combine_1; var var_1; class Group; histogram var_1 / midpoints = (0 to 100 by 5) nrow = 3 odstitle = "VAR" normal (color = red) kernel cfill = ltgray ctext = blue nohlabel vscale = percent; inset N = 'N' MEDIAN (8.2) MEAN (8.2) STD = 'Standard Deviation'(8.3) / position = ne; ods select histogram; run; title 'var_1'; proc boxplot data = combine_1; plot var_1*group / nohlabel; label var_1 = 'VAR'; *inset min max / pos = topoff; insetgroup min (5.2) max (5.2) mean (5.2) stddev (5.2) / header = 'Stats by VAR'; run; ods _all_ close; ods preferences;   Thanks, C ... View more

Hi All,   Is that possible to use proc template in macro? I use a proc templat in SAS to combine 6 figures in to one page, that's good. But I have lots of variables, which means I have to run this proc template many times, but I don't want to change every variable name after I run out proc template once.  Someone can give me an  example to run proc template but use macro? I google, but didn't find a good example.   Thanks, C ... View more

Hi All,   Is there a way to combine two tables but with different format in SAS? For example: I have two tables Variable_1 and Variable_2, ABC columns are character format, DEF columns are PVALUE format, I want  to make a table as shown below, is that possible? I use: data want; set Variable_1 Variable_2; run; But can't output the same table as below.   Variable_1 A B C   Child Young Teen Variable_2 D E F   0.3 0.4 0.5   Thanks, C ... View more

Hi All,   What should I do if I want to extract all the observations have same value but may in different two rows in two columns. For example: Col_1      Col_2 1              1 2              2 3              3 4              5 5              6 6              7 7              9 8              10 The red color records are observations I want to extract. I am using SAS 9.4.    Any idea?   Thanks, C   ... View more

Hi All,   I have a column, the format is $6. , I want to convert it into Date9. , I tried to used code below, but it doesn't work.  data date_type; set date_type; if ParentCaregiverCompleted = "#NULL!" then delete; PCC = put(input(ParentCaregiverCompleted||'01', yymmdd8.),date9.); run;   The data in exced is below: ParentCaregiverCompleted 05/19/15 11/05/14 but the data in my sas dataset is below: ParentCaregiverCompleted 42143 41948   Any help to solve this?   Thanks, C   ... View more