[R] Quartile regression question

[Philippe Grosjean]

Hello,

Look at package quantreg.

Philippe Grosjean


Ranney, Steven wrote:
> I have data that looks like
> 
> lake,loglength,logweight
> 1,2.369215857,1.929418926
> 1,2.426511261,2.230448921
> 1,2.434568904,2.298853076
> 1,2.437750563,2.298853076
> 1,2.442479769,2.230448921
> 1,2.445604203,2.356025857
> ...
> 102,2.722633923,3.310268367
> 102,2.781755375,3.502153893
> 102,2.836324116,3.683407299
> 102,2.802773725,3.583312152
> 102,2.790285164,3.546419267
> 102,2.806179974,3.599118565
> 102,2.716837723,3.316180099
> 
> 
> I can regress log weight on log length simply enough, but how would I model the third quartile of log weights?  In other words, rather than finding a 2nd quartile (or 50th percentile) regression line, 
> 
> e.g., mod=lm(logweight~loglength)
> 
> can R find a 75th percentile line?  Further, since my data is lake>1, is there a way to run 3rd quartile regressions on each lake?  I would imagine that regressing each population would require some call of the subset function, but I cannot figure out how to call it.
> 
> Thanks in advance, 
> 
> SR 
> 
> Steven H. Ranney
> Graduate Research Assistant (Ph.D)
> USGS Montana Cooperative Fishery Research Unit
> Montana State University
> PO Box 173460
> Bozeman, MT 59717-3460
> 
> phone: (406) 994-6643
> fax:   (406) 994-7479
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>