[EXTERNAL] Re: modeling variation along a gradient
You might want to consider modeling multiple quantiles in a quantile regression (package quantreg) as a more general, highly flexible way of modeling variation in a response (y) as some function of predictors (X) without having to specify a distributional form for the conditional response. Lots of options available in quantreg package with many examples in the vignettes provided by Roger Koenker. Brian Brian S. Cade, PhD U. S. Geological Survey Fort Collins Science Center 2150 Centre Ave., Bldg. C Fort Collins, CO 80526-8818 email: cadeb at usgs.gov<mailto:brian_cade at usgs.gov> tel: 970 226-9326
From: R-sig-ecology <r-sig-ecology-bounces at r-project.org> on behalf of Stratford, Jeff <jeffrey.stratford at wilkes.edu>
Sent: Thursday, October 27, 2022 6:26 AM
To: Zolt?n Botta-Duk?t <botta-dukat.zoltan at ecolres.hu>
Cc: r-sig-ecology at r-project.org <r-sig-ecology at r-project.org>
Subject: [EXTERNAL] Re: [R-sig-eco] modeling variation along a gradient
Sent: Thursday, October 27, 2022 6:26 AM
To: Zolt?n Botta-Duk?t <botta-dukat.zoltan at ecolres.hu>
Cc: r-sig-ecology at r-project.org <r-sig-ecology at r-project.org>
Subject: [EXTERNAL] Re: [R-sig-eco] modeling variation along a gradient
This email has been received from outside of DOI - Use caution before clicking on links, opening attachments, or responding. Hi Zoltan, The goal is to model variation per se not the mean (so a function like var = a + Bx instead of y = a + Bx). Thanks for the suggestion for glmmTMB - I'll look into it. Best, Jeff ******************************************************** Jeffrey A. Stratford, PhD. Department of Biology and Earth System Sciences 84 W South Street Wilkes University, PA 18766 USA https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsites.google.com%2Fa%2Fwilkes.edu%2Fstratford%2F&data=05%7C01%7Ccadeb%40usgs.gov%7C52d259635285400e393a08dab816a808%7C0693b5ba4b184d7b9341f32f400a5494%7C0%7C0%7C638024704700574699%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=KP2cPwXKF6ZFUIPpKbniu4hobHuwh0dGVaUkfUxL%2F9s%3D&reserved=0 ******************************************************** On Thu, Oct 27, 2022 at 2:30 AM Zolt?n Botta-Duk?t < botta-dukat.zoltan at ecolres.hu> wrote: > Dear Jeff, > > I'm not sure if you want to model mean, with accounting variation, or to > modelling the change of variation along gradient. Note that in most > distribution (Gaussian is an exception) mean and variation are not > independent, and often variation increases with increasing mean. So > choosing the correct distribution may solve the problem when the aim is > modelling the mean. If not, or the aim is modelling variation, by > glmmTMB you can fit models both to mean and variance. > > Best wishes, > > Zoltan > > 2022. 10. 27. 2:25 keltez?ssel, Stratford, Jeff ?rta: > > Hi all, > > > > I have several data sets that relate some ecological phenomenon to a > > gradient (e.g., avian species richness and urbanization, weevil > prevalence > > in acorns and latitude - see > https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Flinkprotect.cudasvc.com%2Furl%3Fa%3Dhttps%253a%252f%252fwww.mdpi.com%252f1424-2818%252f13%252f7%252f303%26c%3DE%2C1%2C0cKp2p1DFujYtllwpy1B4D8YXPODGboC24UE7K8-WmmQA8SHSMN_KuZ3nqTTyQLyp7oD4zyt2JQqLPfjwQ2yFskgIviy3du5WZOU07vtA2RpKPaH%26typo%3D1&data=05%7C01%7Ccadeb%40usgs.gov%7C52d259635285400e393a08dab816a808%7C0693b5ba4b184d7b9341f32f400a5494%7C0%7C0%7C638024704700574699%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=URwpGDL71%2Fw1D3LK2mL4nfhi%2Fo1pY3%2B46izBWix%2BLbg%3D&reserved=0). > > > Most > > of these data sets have more variation on one end than the other. I have > > considered quantile regression but this seems inadequate. I would like to > > model variation per se along a gradient. > > > > Is there an R function that models sd or variance along a gradient. I was > > thinking that it would require a moving window of some sorts. > > > > Any suggestions would be appreciated. > > > > Thanks, > > > > Jeff > > ******************************************************** > > Jeffrey A. Stratford, PhD. > > Department of Biology and Earth System Sciences > > 84 W South Street > > Wilkes University, PA 18766 USA > > https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsites.google.com%2Fa%2Fwilkes.edu%2Fstratford%2F&data=05%7C01%7Ccadeb%40usgs.gov%7C52d259635285400e393a08dab816a808%7C0693b5ba4b184d7b9341f32f400a5494%7C0%7C0%7C638024704700574699%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=KP2cPwXKF6ZFUIPpKbniu4hobHuwh0dGVaUkfUxL%2F9s%3D&reserved=0 > > ******************************************************** > > > > [[alternative HTML version deleted]] > > > > _______________________________________________ > > R-sig-ecology mailing list > > R-sig-ecology at r-project.org > > https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-ecology&data=05%7C01%7Ccadeb%40usgs.gov%7C52d259635285400e393a08dab816a808%7C0693b5ba4b184d7b9341f32f400a5494%7C0%7C0%7C638024704700574699%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=T3aqvnyu%2F1N%2BtHJyXyP%2FnOL%2FwGT%2FQwKzn%2BMLBzpdPuM%3D&reserved=0 > > -- > Botta-Duk?t Zolt?n Zolt?n Botta-Duk?t > tudom?nyos tan?csad? scientific advisor > ?kol?giai Kutat?k?zpont Centre for Ecological Research > ?kol?giai ?s Botanikai Int?zet Institute of Ecology and Botany > > _______________________________________________ > R-sig-ecology mailing list > R-sig-ecology at r-project.org > https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-ecology&data=05%7C01%7Ccadeb%40usgs.gov%7C52d259635285400e393a08dab816a808%7C0693b5ba4b184d7b9341f32f400a5494%7C0%7C0%7C638024704700730933%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=yZmqQiRPLhtyjslk9wh4Fgv0NqtQLR%2BrFg8%2BeLfm%2Fa0%3D&reserved=0 > [[alternative HTML version deleted]] _______________________________________________ R-sig-ecology mailing list R-sig-ecology at r-project.org https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-ecology&data=05%7C01%7Ccadeb%40usgs.gov%7C52d259635285400e393a08dab816a808%7C0693b5ba4b184d7b9341f32f400a5494%7C0%7C0%7C638024704700730933%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=yZmqQiRPLhtyjslk9wh4Fgv0NqtQLR%2BrFg8%2BeLfm%2Fa0%3D&reserved=0