### [R-sig-eco] Using rq() for least absolute deviation regression

I've seen several websites say that the function rq() from the package quantreg can be used to do least absolute deviation regression. How do you go about doing this and what's the connection between quantile regression and LAD? (I'm very new to the former topic.) Thanks, Jane -- -

### Re: [R-sig-eco] Using rq() for least absolute deviation regression

The short answer is that what you seem to want is the rq() default with tau not specified. (Default is tau=.5). In general rq() minimizes a sum of weighted absolute residuals. The weights depend on tau (the conditional quantile of interest), and turn out to be equal with tau = 0.5, i.e.,

### Re: [R-sig-eco] subsetting data in R

If this isn't already answered: I don't quite understand the question: what do you mean by do a complete data set from an object in R? What do you mean by the subsetting is dangerous ... as you need to specify the levels for all your factors again? (What do your 3000 columns of data

### Re: [R-sig-eco] The final result of TWINSPAN

Dear List, Earlier this year on an (undoubtedly ill-advised) lark I coded up an R version of TWINSPAN. It's far from a polished package at this point, but the code does run. One of the interesting features is that you can partition a PCO or NMDS in addition to the traditional CA. To be

### Re: [R-sig-eco] subsetting data in R

Thank you very much Ben. I was doing an analysis of indicator species with the subset data and the other levels were still in my subset data and the analysis was considering them in the analysis. My 3000 columns are plant species presence/absence type of data. Best, Manuel On 26/04/2011

### Re: [R-sig-eco] The final result of TWINSPAN

On 27/04/11 00:40 AM, Dave Roberts dvr...@ecology.msu.montana.edu wrote: Earlier this year on an (undoubtedly ill-advised) lark I coded up an R version of TWINSPAN. It's far from a polished package at this point, but the code does run. One of the interesting features is that you can