[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

*To*: bsouthey@bigfoot.com*Subject*: Re: Calcuation of STND residuals*From*: Arthur Gilmour <gilmoua@apollo.agric.nsw.gov.au>*Date*: Thu, 20 Apr 2000 10:17:24 +1000 (EST)*Cc*: asreml@chiswick.anprod.csiro.au*Reply-To*: Arthur Gilmour <gilmoua@apollo.agric.nsw.gov.au>*Sender*: asreml-owner@lamb.chiswick.anprod.csiro.au

Dear Bruce, > X-Authentication-Warning: lamb.chiswick.anprod.csiro.au: petidomo set sender to asreml-owner@lamb.chiswick.anprod.csiro.au using -f > Date: Wed, 19 Apr 2000 13:21:22 -0500 > From: Bruce Southey <southey@uiuc.edu> > MIME-Version: 1.0 Dear Bruce, It is negligent of me to explain how the STND RES are calculated. I wrote this bit of code long before thinking of calculating the HAT matrix diagonal. I wanted a robust estimate of the standard error of the actual values so my criterion is sum(abs(res))/count/0.8 In this example S> sum(abs(yht$Res))/21/.8 [1] 1.2054 S> sum(abs(yht$Res))/21/.8*3.3 [1] 3.9777 S> yht$Res[9]/1.2054 [1] 3.3184 S> 21 is the count because I ignore the 3 zero residuals on the [false in this case] assumption that they relate to missing values. You are write in that I'm using 3.3 of the standard value. Now that I calculate the hat diagonal, I could calculate the standardized residual if I have the information available at the right time. I'll think about doing that. In the meantime, it is not difficult to calc the Stand Res from what is in the yht file S> yht$Res/sqrt(1-yht$Hat) [1] -0.41775 -0.41775 0.69625 0.69625 0.00000 0.00000 1.07180 -1.07180 [9] 4.28722 -0.41775 -0.41775 -0.41775 -1.07180 -1.07180 1.07180 2.87706 [17] -0.95902 -0.95902 0.69625 -1.07180 -1.07180 0.00000 -1.07180 -0.95902 Arthur > To: "asreml@chiswick.anprod.csiro.au" <asreml@chiswick.anprod.csiro.au> > Subject: Calcuation of STND residuals > > Hi, > Could someone please tell me how the STND RES are calculated? > I can not get these residuals from the .yht file. > The attached is a simple example with a fixed effects only model. > > I ran the analysis with and without the data point - S^2 is the variance > with it and S(i)^2 is the variance without. Res and Hat are the > residuals and Hat value from the .yht file. > > As per pg 376-377 of McCullagh and Nelder > Stand Res was calculated as Res /(sqrt(1-Hat)) > Student Res was calculated as Stand Res/S > Del Stud Res was calculated as Student Res * S / S(i) > > S^2 S(i)^2 Res Hat Stand Res Student Res > Del Stud Res > 1.55357 0.758523 4 0.1295 4.287218357 3.439617822 7.044858309 > > The value from the .res file is 3.32. > > Also, in at least this version of ASREML, potential outliers are those > with STND RES of >3.3 rather than the 3.5 said on page 115 of the March > 21, 2000 manual. > > Thanks > Bruce <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> Arthur Gilmour PhD mailto:Arthur.Gilmour@agric.nsw.gov.au Principal Research Scientist (Biometrics) fax: <61> 2 6391 3899 NSW Agriculture <61> 2 6391 3922 Orange Agricultural Institute telephone work: <61> 2 6391 3815 Forest Rd, ORANGE, 2800, AUSTRALIA home: <61> 2 6362 0046 ASREML is still free by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk or point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/ To join the asreml discussion list, send the message subscribe mailto:asreml-request@chiswick.anprod.CSIRO.au To send messages to the list, mailto:asreml@chiswick.anprod.CSIRO.au Asreml list archive: http://www.chiswick.anprod.csiro.au/lists/asreml <> <> <> <> <> <> <> "Why do you seek the living among the dead? He is not here but is risen! Remember how He spoke to you while He was still in Galilee, saying 'The Son of Man must be delivered into the hands of sinful men, and be crucified, and THE THIRD DAY RISE AGAIN.'" angels, Luke 24:5-7 Many claim to be guides in life. Only Jesus validated His claim by living according to His own teaching and declaring His death and resurrection before the event. <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> -- Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml

- Prev by Date:
**Calcuation of STND residuals** - Next by Date:
**Re: Genetic groups..** - Prev by thread:
**Calcuation of STND residuals** - Next by thread:
**Computing environments for ASREML** - Index(es):