Re: Is my model right for survival rate?

Hi,
>I strongly suggest you also fit a sire model, and compare the results,
>as the animal model can give biassed estimates of the variance with
>binomial data.

Please be very careful of this because the sire model, by definition, is overdispersed compared to the animal model. Unlike assuming normality, the sire and animal models are not linearly equivalent models.

Bruce

---- Original message ----
>Date: Fri, 14 Sep 2007 14:16:23 +1000
>From: arthur.gilmour_at_DPI.NSW.GOV.AU
>Subject: Re: Is my model right for survival rate?
>To: ASREML-L_at_AGRIC.NSW.GOV.AU
>
>Dear luan sheng
>
>re:
>I am the beginner of asreml. I am estimating the variance of survival
>rate which is from the challenge test data of turbot. Turbot is one of
>marine fishies.
>this is my model:
>survival !bin !probit !wt=0 ~ mu !r animal fullsib 0.2
>Is the model right for this analysis? which is right for survival rate
>logit or probit link function?
>The format of data file is just like this:
>animal Sire Dam fullsib survival (binary,0 for death, 1 for survival)
>20070028 2006041 20060159 3 0
>20070029 2006041 20060159 3 0
>20070030 2006041 20060159 3 0
>20070031 2006041 20060159 3 0
>
>I strongly suggest you also fit a sire model, and compare the results,
>as the animal model can give biassed estimates of the variance with
>binomial data.
>In this case, you appear to have large families so it might be fine.
>
>survival !bin !probit !wt=0 ~ mu !r Sire fullsib 0.2
>
>I cannot tell from the data whether Sire and Dam provide the same grouping
>of records
>(i.e. you only have full sib families). If you have half sib families,
>then also\
>fit Dam.
>
>
>As a new user, you should try several forms of the models to get a feel
>for how models compare. e.g. You can also fit the models as if the
>traits were
>normal. Your model is simple, and appears to have an incidence around 50%
>so a Normal analysis should be consistent with the binary analysis.
>
>For binary data, you can omit the '!wt=0' qualifier as 0 is the default
>setting.
>
>
>
>
>May Jesus Christ be gracious to you,
>
>Arthur Gilmour, His servant .
>
>Mixed model regression mapping for QTL detection in experimental crosses.
>Computational Statistics and Data Analysis 51:3749-3764 now available at
>http://dx.doi.org/10.1016/j.csda.2006.12.031
>
>Profile: http://www.dpi.nsw.gov.au/reader/17263
>Personal website: http://www.cargovale.com.au/
>
>mailto:Arthur.Gilmour_at_dpi.nsw.gov.au, arthur_at_cargovale.com.au
>Principal Research Scientist (Biometrics)
>NSW Department of Primary Industries
>Orange Agricultural Institute, Forest Rd, ORANGE, 2800, AUSTRALIA
>
>fax: 02 6391 3899; 02 6391 3922 Australia +61
>telephone work: 02 6391 3815; home: 02 6364 3288; mobile: 0438 251 426
>
>ASREML 2 is now available from http://www.VSNi.co.uk/products/asreml
>The ASReml discussion group is at ASREML-L_at_dpi.nsw.gov.au
>To join it, mailto:arthur.gilmour_at_dpi.nsw.gov.au
>Archives are at
>https://gatekeeper.dpi.nsw.gov.au/Listserv/archives/asreml-l.html
>Cookbook: http://uncronopio.org/ASReml
>
>Proposed travel:
>Leave 10 Aug - 15 Aug
>AAABG 24-27 September Armidale
>Leave 1Oct to 12 Oct
>
> <><><><><><><><><><><><><><><><><><><><><><><><><>
>
>This message is intended for the addressee named and may contain
>confidential information. If you are not the intended recipient or
>received it in error, please delete the message and notify sender. Views
>expressed are those of the individual sender and are not necessarily the
>views of their organisation.
Received on Tue Sep 14 2007 - 08:30:13 EST