About MikeZdeb

MikeZdeb · ‎12-22-2011

hi ... another idea ... data new; infile 'z:\test.txt' end=last; input a b c; if ^last; run;

MikeZdeb · ‎12-20-2011

hi ... it's not that a different part of the queue is used LAG gives you the value of a variable from the last time the LAG function was executed, NOT the value of the variable in the previous observation (that's what Tom told you) so ... if you use IF-THEN-ELSE without the statements ... LAGA = LAG(A); LAGB = LAG(B); then LAG(A) and/or LAG(B) may not get executed during each pass through the data step

MikeZdeb · ‎12-20-2011

hi ... you can have a LAG function within a conditional statement that does execute in every iteration of the data step if you use an IFN function instead of your IF-THEN-ELSE statements... data test3; retain f 1; set test; f = ifn(a ne lag(a) or missing(lag(a)) , 1 , ifn(a eq lag(a) and b gt lag(b) , f+1 , f )); run; the above gives the same result as one of your data step, just modified slightly by moving the LAG statements after the SET statement ... DATA TEST1; RETAIN F ( 1) ; SET TEST; laga = lag(a); lagb=lag(b); IF A^=laga OR laga =. THEN do; f=1;end; ELSE IF A=laga AND b>lagb THEN do; f=f+1 ; end; RUN; also fyi ... you don't need the DO-END stuff or the parentheses in the RETAIN and you can use a SUM statement ... DATA TEST1; RETAIN F 1 ; SET TEST; laga = lag(a); lagb=lag(b); IF A^=laga OR laga =. THEN f=1; ELSE IF A=laga AND b>lagb THEN f+1; RUN; and ... since in the first pass through the data step, laga is missing, you can skip the RETAIN statement that gives F an intial value of 1 since F will be assigned a 1 in the first pass and the sum statement (F+1) is an "implied RETAIN" for F ps you can read more about LAG and IF in Howard Schreier's paper ... "Conditional Lags Don't Have to be Treacherous" http://www.howles.com/saspapers/CC33.pdf

MikeZdeb · ‎12-20-2011

hi ... one of the attributes of a a SYMBOL is COLOR and if you do not specify a color for a symbol, SAS/GRAPH cycles though the color list for a given symbol before using the next symbol so ... I added a color to your symbol statements and ... I added a data step to make some fake data in the range of your axis statements and produced the attached plot and ... I made a few changes to some of your SAS code #1 not sure why you are using a SAS stroke font (swiss) when TrueType fonts (Calibri, whatever) might be available and look better #2 raised the pixel count ... looks better in bit-map output (GIF) #3 used a TrueType font (Wingdings) as the symbol value ... if you have access to TrueType fonts try ... goptions reset-all; proc gfont name="wingdings" nibuild showroman; run; quit; character '6c'x is a DOT #4 added a LEGEND definition to get rid of the default label (PLOT1) #5 added some "white space" around the chart with the LS option for title/footnote here's the code and the result is attached ... data plot_final; do _n_ = 1 to 50; Single_Coefficient = 100 + ceil(900*ranuni(1234)); Coverage_Plus1 = 20 + ceil(80*ranuni(1234)); p_1 = 20 + ceil(80*ranuni(1234)); output; end; run; proc sort data=plot_final; by Single_Coefficient; run; filename grafout "z:\chart_new.gif"; goptions reset=global device=gif gsfname=grafout gsfmode=replace xpixels=1500 ypixels=1000 gunit=pct border cback=white colors=( blue black)ftitle="calibri/bo" ftext="calibri" htitle=3 htext=2; axis1 order=(100 to 1000 by 100) offset=(3,3) color=black major=(height=1 width=1) minor=none width=1; axis2 label=(angle=270 rotate=90 height=2 'Coverage') order=(20 to 100 by 10) offset=(3,3) color=black major=(height=1 width=1) minor=none width=1; legend1 label=none; symbol1 height=0.5 f="wingdings" value='6c'x c=blue; symbol2 f="wingdings" value='6c'x interpol=join line=1 height=1.3 c=blue; title1 'CURVE CHART' ls=2; title2 a=90 ls=1; title3 a=-90 ls=1; footnote1 ls=0.5; proc gplot data=plot_final; plot Coverage_Plus1*Single_Coefficient p_1*Single_Coefficient /overlay haxis=axis1 vaxis=axis2 caxis=black cframe=white legend=legend1; run; quit;

MikeZdeb · ‎12-13-2011

hi ... it's OK that the two variables are character variables (in the example I posted, the group variable SEX is a character variable) you do not need X, Y, and Z ... if you look at the exmaple I posted, I just have MIDPOINT, GROUP, and X and get annotate labels placed on the chart at this point, I think that the only way you'll get some meaningful help with this is (if the data set is small enough) to attach you data set to a postinh so others can give it a try

MikeZdeb · ‎12-12-2011

hi ... first, you'd get a LOT more attention for this question on the SAS/GRAPH AND ODS GRAPHICS community page second, I think you need a value for X in the annotate data set ... where does the label go along the x-axis you have a value for midpoint and group and they position the label on the y-axis on a horizontal bar chart your SUMVAR is VALORE, so have you tried adding x = valore; to the data step that creates the annotate data set here's a stripped down version of your SAS code that uses SASHELP.CLASS to produce a chart ... all I added to thet the annotate data set is the variable X it produced the attached chart (doesn't show anything useful but it did work) proc summary data=sashelp.class nway; class sex age; var weight; output out=stats sum=; run; data anlabels2; length text $ 20 ; retain function 'label' style '"Albany AMT/bo"' when 'a' xsys ysys '2' hsys '3' size 3 position '6' color 'black'; set stats; midpoint=age; group=sex; text=cat(weight); x = weight; run; goptions reset=all; proc gchart data=sashelp.class; hbar3d age / discrete sumvar=weight SHAPE=cylinder patternid=midpoint group=sex nostats annotate=anlabels2; run; quit; ps not sure what all your annotate errors mean

MikeZdeb · ‎12-11-2011

hi ... you could also use Art's suggestion in one step ... proc sql; create table want as select *, (year gt year(today())-10) as group from have order by group, year; quit;

MikeZdeb · ‎12-07-2011

hi ... if the data set is in the order you show and that output is all you want using data _null_ ... data _null_; file print; if _n_ eq 1 then put ' YEAR PRICE1 PRICE2 PRICE3'; set x; if year ne 2011 then do; put (year price1-price3) (5. 3*8.); t1+price1; t2+price2; t3+price3; end; else do; put / 'TOTAL' (t1-t3) (8.); put (year price1-price3) (5. 3*8.); t1+price1; t2+price2; t3+price3; put 'TOTAL' (t1-t3) (8.); end; run;

MikeZdeb · ‎12-07-2011

hi ... similar to Tom's idea ... data x (drop=temp:); infile datalines truncover; input @1 @'{1:' field1 $4. @1 @'{2:' field2 $5. @1 @'{3:' temp3 $100. @1 @'{4:' temp4 $10. ; if ^missing(temp3) then field3 = substr(temp3,1,find(temp3,'}}')); field4 = input(compress(temp4,'.','kd'),10.); datalines; {1:ABCD}{2:10234}{4:40.45} {1:TRHG}{2:38976}{3:{H132}}{4:567.43} {1:LKMN}{2:23109}{4:10432.67} {1:HGKO}{2:19876}{3:{J987}{F102}}{4:4320.12} {1:DUHN}{2:23098}{3:{G123}{H987}{J234}}{4:345.23} ; run;

MikeZdeb · ‎12-03-2011

hi ,,, using Ksharp's data and suggestion about PROC MEANS (or SUMMARY) and the IDGROUP option ... * find maximum group size; proc sql noprint; select max(obs) into :obs from (select count(*) as obs from have group by id); quit; * transpose three variables at one time; proc summary nway data=have missing; class id; output out = want (drop=_:) idgroup(out[&obs](visit_date dose location)=); run; for more info, you can also look at .. "Transposing Data Using PROC SUMMARY'S IDGROUP Option" http://support.sas.com/resources/papers/proceedings10/102-2010.pdf

MikeZdeb · ‎12-03-2011

hi ... nicer without the "else..." but unfortunately that generates a id=0 (rather than missing) for the missing num rows given how the sum statement works (with an initial value of zero) without the "else" ... but retain is also an extra statement data y; retain id .; set x; by num; if ^missing(num) then id+first.num; run; the use of +first.<var> for a group counter comes from posts I saw ... and remembered, that's the hard part ... by Howard Schreier a long time ago Howard is the elegant code master

MikeZdeb · ‎11-29-2011

hi ... this gives the ouput you specified I added an extra set of observations where the (a, -a) pair crosses values of NOSINID ... data have; input Nosinid ndep dep; cards; 1 1 100 1 2 -100 1 3 100 2 1 500 2 2 -500 2 3 500 2 4 -500 2 5 500 2 6 500 3 1 1000 3 2 -500 3 3 500 4 1 -500 4 2 100 4 3 100 ; run; data want (drop=n d); do j=1 to obs; set have nobs=obs point=j; k = j + 1; set have (keep=nosinid dep rename=(nosinid=n dep=d)) point = k; if (nosinid eq n) and (dep gt 0) and (dep+d eq 0) then j + 1; else output; end; stop; run; Nosinid ndep dep 1 3 100 2 5 500 2 6 500 3 1 1000 3 2 -500 3 3 500 4 1 -500 4 2 100 4 3 100

MikeZdeb · ‎11-29-2011

hi ... you already got an answer, but here's another suggestion ... data y; set x; by num; if ^missing(num) then id+first.num; else id = .; run;

MikeZdeb · ‎11-25-2011

hi ... no different than what happens a data step ... data males; set sashelp.class (keep=sex name age rename=(sex=gender)); where gender eq 'M'; run; proc sql; create table males as select sex as gender, name age from sashelp.class where gender eq 'M'; quit;

MikeZdeb · ‎11-23-2011

hi .. lots of answers posted already, but here's another (it uses TRANSPOSE like a few others) that uses the SMALLEST function ... * some fake data (10 random ids, 10 numeric variables); data have; array x(10); do j = 1 to 10; id = put(ceil(1000*ranuni(999)),z4.); do k = 1 to 10; x(k) = ceil(100*ranuni(999)); x(k) = ifn(ranuni(999) le .3, x(k)*-1, x(k)); end; output; end; drop j k; run; proc transpose data=have out=want; var _numeric_; run; * use SMALLEST (use COUNTC to determine the number of values to skip); data want; set want; smallest_positive = smallest(countc(cats(of _numeric_),'-') + 1, of _numeric_); which_obs = whichn(smallest_positive, of _numeric_); set have (keep=id) point=which_obs; keep sm: id _name_; run; results ... smallest_ _NAME_ positive id x1 12 0778 x2 14 0778 x3 16 0423 x4 13 0609 x5 16 0031 x6 8 0031 x7 8 0352 x8 21 0750 x9 15 0750 x10 3 0491

Online Status	Offline
Date Last Visited	‎06-28-2018 10:46 AM

Re: Create variables as per separators in string

Re: How to create sub-totals ?

Re: Running a function for each unique value

Re: Aggregating dataset by date, including missing dates

Re: Aggregating dataset by date, including missing dates

Re: Aggregating dataset by date, including missing dates

Re: Aggregating dataset by date, including missing dates

Re: How to write generic libname to multiple databases?

Re: delete/retein observations by frequency (n) of variables

Re: delete/retein observations by frequency (n) of variables

Re: is there simpler way to convert char flag to numeric number

Re: INPUT function requires a character argument

Re: Renaming all variables in a table

Re: Report on missing values of each coloumn

Re: Remove the last character of my string

Re: Proc GMap: Caption and custom legend

Re: Saving PROC GPLOT graph to folder

Re: re: Catx Function Error

How to calculate geodesic distance in SAS

Re: Excluding last row when importing a file

data step execution

Re: data step execution

Re: Plotting two curves in a single Graph Using Proc GPLOT option Over...

Re: Proc gchart and annotate: problem in all observations

Re: Proc gchart and annotate: problem in all observations

Re: Suggestion on grouping

Re: Summing Based on condition for Year

Re: Help - Infile Multiple variable Length Fields

Re: Transposing a dataset

Re: Generating ID but don't want to have an id for blank value rows

Re: Deleting consecutive values

Re: Generating ID but don't want to have an id for blank value rows

Re: SQL where statement using alias column

Re: The smallest positive value for each variable