From: IN%"Gregor.Gorjanc@bfro.uni-lj.si" "Gorjanc Gregor" 17-APR-2006 08:31:17.71 To: IN%"Applied-ethology@sask.usask.ca" CC: Subj: Help with analysis of covariance structure (correlations) between animal behaviour Hello! This is a data analysis (statistics) question and I hope someone here can provide me relevant answer or point me to relevant sources. I am analysing data on behaviour of animals. I have a factor at 1-level and I also added animal indicator as 2-level of my model, as there is considerable variation between animals. Very often one assumes that 2-level units are independent, but in my case this might not be the case, since animals within one level of a factor were in contact and might influence on each other. I could model this dependency as say multivariate normal with particular covariance matrix for 2-level within each factor, but I have not came accros any such examples. I a bit skeptic in amount of information to estimate "many" covariance components. For now compound symmetry, unstructured and Toeplitz (in SAS parlance[1]) structures for covariance matrix have come to my mind fot this task. Does anyone have any experience in this type of models? Any suggestions, comments, pointer to relevant literature are more than welcome. Thanks! [1]http://v8doc.sas.com/sashtml/stat/chap41/sect20.htm#mixedspcovstruct Lep pozdrav / With regards, Gregor Gorjanc ---------------------------------------------------------------------- University of Ljubljana Biotechnical Faculty URI: http://www.bfro.uni-lj.si/MR/ggorjan Zootechnical Department mail: gregor.gorjanc bfro.uni-lj.si Groblje 3 tel: +386 (0)1 72 17 861 SI-1230 Domzale fax: +386 (0)1 72 17 888 Slovenia, Europe ---------------------------------------------------------------------- "One must learn by doing the thing; for though you think you know it, you have no certainty until you try." Sophocles ~ 450 B.C. ---------------------------------------------------------------------- From: IN%"nabilbrandl@privat.dk" "Nabil Brandl" 17-APR-2006 12:15:14.36 To: IN%"Applied-ethology@sask.usask.ca" "Applied Ethology" CC: Subj: Covariance vs regression analysis Here is a comparison between regression analysis and covariace analysis http://www.uku.fi/~mauranen/advbis/advbis4.htm Best regards Nabil Brandl Fax 45-86624697 Mobile: 45-40895962 e-mail: NabilBrandl@tiscali.dk e-mail: nabilbrandl@privat.dk e-mail: nabilbrandl@orangenet.dk e-mail: nabilbrandl1@excite.com http://nabilnabil.homestead.com http://home20.inet.tele.dk/nabilbrandl/index.html From: IN%"Gregor.Gorjanc@bfro.uni-lj.si" "Gorjanc Gregor" 17-APR-2006 15:15:19.63 To: IN%"Applied-ethology@sask.usask.ca" CC: Subj: RE: Covariance vs regression analysis Nabil Brandl wrote: >> Here is a comparison between regression analysis and covariace analysis >> http://www.uku.fi/~mauranen/advbis/advbis4.htm Thank you Nabil. Unfortunatelly, this does not solve my problem. I would like to see any analysis with linear or generalized linear (possibly also mixed) models that account for covariance between animals. Lep pozdrav / With regards, Gregor Gorjanc ---------------------------------------------------------------------- University of Ljubljana Biotechnical Faculty URI: http://www.bfro.uni-lj.si/MR/ggorjan Zootechnical Department mail: gregor.gorjanc bfro.uni-lj.si Groblje 3 tel: +386 (0)1 72 17 861 SI-1230 Domzale fax: +386 (0)1 72 17 888 Slovenia, Europe ---------------------------------------------------------------------- "One must learn by doing the thing; for though you think you know it, you have no certainty until you try." Sophocles ~ 450 B.C. ---------------------------------------------------------------------- From: IN%"nabilbrandl@privat.dk" "Nabil Brandl" 17-APR-2006 21:31:20.23 To: IN%"Gregor.Gorjanc@bfro.uni-lj.si" "Gorjanc Gregor", IN%"Applied-ethology@sask.usask.ca" CC: Subj: RE: Covariance vs regression analysis Here is an article, which describe generalized and mixed model. I hope it will help you to solve your problem. Some applications of generalized linear latent and mixed models in epidemiology: Repeated measures, measurement error and multilevel modeling: http://www.gllamm.org/epi7.pdf Best regards Nabil Brandl Fax 45-86624697 Mobile: 45-40895962 e-mail: NabilBrandl@tiscali.dk e-mail: nabilbrandl@privat.dk e-mail: nabilbrandl@orangenet.dk e-mail: nabilbrandl1@excite.com http://nabilnabil.homestead.com http://home20.inet.tele.dk/nabilbrandl/index.html ----- Original Message ----- From: "Gorjanc Gregor" To: Sent: Monday, April 17, 2006 11:16 PM Subject: Re: Covariance vs regression analysis > Nabil Brandl wrote: > >>> Here is a comparison between regression analysis and covariace analysis >>> http://www.uku.fi/~mauranen/advbis/advbis4.htm > > Thank you Nabil. Unfortunatelly, this does not solve my problem. I would > like to see any analysis with linear or generalized linear (possibly > also mixed) models that account for covariance between animals. > > Lep pozdrav / With regards, > Gregor Gorjanc > > ---------------------------------------------------------------------- > University of Ljubljana > Biotechnical Faculty URI: http://www.bfro.uni-lj.si/MR/ggorjan > Zootechnical Department mail: gregor.gorjanc bfro.uni-lj.si > Groblje 3 tel: +386 (0)1 72 17 861 > SI-1230 Domzale fax: +386 (0)1 72 17 888 > Slovenia, Europe > ---------------------------------------------------------------------- > "One must learn by doing the thing; for though you think you know it, > you have no certainty until you try." Sophocles ~ 450 B.C. > ---------------------------------------------------------------------- From: IN%"fmartin@vetmed.wsu.edu" "Martin, Francois" 18-APR-2006 14:21:13.14 To: IN%"Applied-ethology@sask.usask.ca" "Ethology Posting" CC: Subj: HAB course Good Afternoon, We have 2 upcoming courses in the Human Animal Bond that may interest = you. Each course is a two hour long presentation which can be accessed = starting April 24. The courses are presented by Dr. Fran=E7ois Martin, = Associate Director of the Center for the Study of Animal Well-Being and = Head of the People-Pet Partnership. These courses are available to = veterinarians, veterinary para professionals and others who are = interested in the bond between animals and humans. Each course will = provide 2 hours of continuing education credit for veterinarians based = on passing an online examination. =20 Please see http://www.vetmed.wsu.edu/ce/hab/ for course details and = registration instructions. Please distribute this information as you see fit. Regards FM Francois Martin, M.A., Ph.D. Associate Director of the Center for the Study of Animal Well-being and Head of the People-Pet Partnership College of Veterinary Medicine Washington State University PO Box 647010 Pullman WA 99164-7010 509 335.4569 fmartin@vetmed.wsu.edu =20 From: IN%"Gregor.Gorjanc@bfro.uni-lj.si" "Gorjanc Gregor" 18-APR-2006 18:14:56.73 To: IN%"Applied-ethology@sask.usask.ca" CC: Subj: Help with analysis of covariance structure (correlations) between animal behaviour Hello! This is a data analysis (statistics) question and I hope someone here can provide me relevant answer or point me to relevant sources. I am analysing data on behaviour of animals. I have a factor at 1-level and I also added animal indicator as 2-level of my model, as there is considerable variation between animals. Very often one assumes that 2-level units are independent, but in my case this might not be the case, since animals within one level of a factor were in contact and might influence on each other. I could model this dependency as say multivariate normal with particular covariance matrix for 2-level within each factor, but I have not came accros any such examples. I a bit skeptic in amount of information to estimate "many" covariance components. For now compound symmetry, unstructured and Toeplitz (in SAS parlance[1]) structures for covariance matrix have come to my mind fot this task. Does anyone have any experience in this type of models? Any suggestions, comments, pointer to relevant literature are more than welcome. Thanks! [1]http://v8doc.sas.com/sashtml/stat/chap41/sect20.htm#mixedspcovstruct Lep pozdrav / With regards, Gregor Gorjanc ---------------------------------------------------------------------- University of Ljubljana Biotechnical Faculty URI: http://www.bfro.uni-lj.si/MR/ggorjan Zootechnical Department mail: gregor.gorjanc bfro.uni-lj.si Groblje 3 tel: +386 (0)1 72 17 861 SI-1230 Domzale fax: +386 (0)1 72 17 888 Slovenia, Europe ---------------------------------------------------------------------- "One must learn by doing the thing; for though you think you know it, you have no certainty until you try." Sophocles ~ 450 B.C. ---------------------------------------------------------------------- From: IN%"Gregor.Gorjanc@bfro.uni-lj.si" "Gorjanc Gregor" 18-APR-2006 18:26:03.90 To: IN%"Applied-ethology@sask.usask.ca" CC: Subj: RE: Covariance vs regression analysis Nabil Brandl wrote: >> Here is a comparison between regression analysis and covariace analysis >> http://www.uku.fi/~mauranen/advbis/advbis4.htm Thank you Nabil. Unfortunatelly, this does not solve my problem. I would like to see any analysis with linear or generalized linear (possibly also mixed) models that account for covariance between animals. Lep pozdrav / With regards, Gregor Gorjanc ---------------------------------------------------------------------- University of Ljubljana Biotechnical Faculty URI: http://www.bfro.uni-lj.si/MR/ggorjan Zootechnical Department mail: gregor.gorjanc bfro.uni-lj.si Groblje 3 tel: +386 (0)1 72 17 861 SI-1230 Domzale fax: +386 (0)1 72 17 888 Slovenia, Europe ---------------------------------------------------------------------- "One must learn by doing the thing; for though you think you know it, you have no certainty until you try." Sophocles ~ 450 B.C. ---------------------------------------------------------------------- From: IN%"fmartin@vetmed.wsu.edu" "Martin, Francois" 18-APR-2006 23:40:14.75 To: IN%"Applied-ethology@sask.usask.ca" "Ethology Posting" CC: Subj: HAB course Good Afternoon, We have 2 upcoming courses in the Human Animal Bond that may interest = you. Each course is a two hour long presentation which can be accessed = starting April 24. The courses are presented by Dr. Fran=E7ois Martin, = Associate Director of the Center for the Study of Animal Well-Being and = Head of the People-Pet Partnership. These courses are available to = veterinarians, veterinary para professionals and others who are = interested in the bond between animals and humans. Each course will = provide 2 hours of continuing education credit for veterinarians based = on passing an online examination. =20 Please see http://www.vetmed.wsu.edu/ce/hab/ for course details and = registration instructions. Please distribute this information as you see fit. Regards FM Francois Martin, M.A., Ph.D. Associate Director of the Center for the Study of Animal Well-being and Head of the People-Pet Partnership College of Veterinary Medicine Washington State University PO Box 647010 Pullman WA 99164-7010 509 335.4569 fmartin@vetmed.wsu.edu =20