|
Thanks a ton Nat for explaining the concept for me. I am getting mean of
lab values and adding and substrate SD from the mean , transposing it to
get one one variable for high low and mean.
To make very number unique I am adding _n_/1000 to it make every number
unique. some thing like this
***mean values variables ***;
proc univariate data=dlb_m noprint;
by armcdn arm visitnum visit ;
var lbstresn ;
output out=dlb_m_out n=n1 median=med1 min=min1 max=max1 mean=mea1
std=std1 q1=q25 q3=q75;
run;
data dlb_m_out;
set dlb_m_out;
if std1 ^= . then mea0 = (mea1 -std1);
if std1 ^= . then mea2 = (mea1 +std1);
run;
proc transpose data = dlb_m_out out = dlb_m_out1 (rename = (col1 = qval));
by armcdn arm visitnum visit mea1;
var mea0 mea1 mea2 ;
run;
Data dlb_m_out1;
set dlb_m_out1;
if qval = . and mea1 ^= . then qval = mea1;
***********Adding the temp number to make every mean unique********;
temp_n = (_n_/1000) ;
if qval ^= . then qval = qval +temp_n;
if mea1 ^= . then mea1 = mea1 +temp_n;
*******************************************************************;
Run;
proc sort data = dlb_m_out1 out = a1 dupOut = aaa Nodupkey ;
by qval ; *no duplicates*;
*by mea1;*no duplicates*;
run;
Even after doing this I am still getting that warning . Is there a better
way to check if they are any duplicates of numbers ? or any way to make all
the mean and qval unique.
WARNING: Greater than three node values for at least one observation in
HILOC interpolation. Mean value used for tick placement.
On Sat, Sep 25, 2010 at 7:53 AM, Nat Wooding <nathani@verizon.net> wrote:
> Kumar
>
> The documentation for HILOC (High, Low, Close ) indicates that if there are
> more than three values for a particular x axis position, then the mean Y
> for
> that x is calculated.
>
> It appears that since SAS is expecting only three values for each X, it
> lets
> you know that you have more than three Y values for at least one situation
> and that it has done the mean calculation. I duplicated the message by
> using
> the sample code from the example and then simply duplicating one of the
> observations.
>
> Nat Wooding
>
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
> SAS_learner
> Sent: Saturday, September 25, 2010 2:14 AM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: PROC GPLOT problem in Log
>
> Hello all ,
>
> I am producing something very similar to
> http://support.sas.com/kb/25/506.html (spaghetti plots using proc gplot in
> sas) . In the Log I am getting
>
> WARNING: Greater than three node values for at least one observation in
> HILOC interpolation. Mean value used for tick placement.
>
> I checked the data and there are 12 visits as the tick marks . This is
> coming in the graph for all the treatments . Individual treatments is good
> with out any warnings ??
>
> Please let me know if the question is not clear I can try to send my
> program with Log for your review.
>
> thanks
> Kumar
>
| 
