R-beta: "smooth sbezier" blew up too much data, need code

John vanVlaanderen (johnvv@hotmail.com)
Mon, 26 Jan 1998 15:26:37 PST


From: "John vanVlaanderen" <johnvv@hotmail.com>
To: info-gnuplot@cs.dartmouth.edu
Subject: R-beta: "smooth sbezier" blew up too much data, need code
Date: Mon, 26 Jan 1998 15:26:37 PST



Hello All,

I am in need of sample code to create gnuplot's sbezier output in 
numerical form.  Below are some details of what I have done.

I just fed 2000 lines of data to the above gplot script and it seg 
faulted, when I cut the data down it down it worked fine.

The results are very good for the purpose, which is to give a 
guestimation as to what fills the spaces between the exceptions.

When used with other highly varing data it makes the lines far more 
readable for the purpose of comparing systems while looking for 
problems.

The smooth sbezier function starts by making "the data monotonic in x"
and then applying the bezier function. What I need to do is take the 
functions out of gnuplot and translate them into perl.  Once outside can 
run the regression script and then I have a single line for visual 
forcasting and also I will be able to run exponential smoothing to 
enhance forcasting.  Another issue with respect to systems forcasting is 
"seasonality as refered to by the census."  Daily, weekly and monthly 
data operations tend to run in cycles.

Any leads to awk, C or perl code would be greatly appreciated.

Thanks in advance, John van V aka johnvv@hotmail.com 









______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._