[GenABEL-dev] Creating a complex coxph model

[Jeff Alstott]

Oops, that should be:

Surv(rep(1, n_rows), column1) ~ column2 + column3 + column2*column3 +
strata(column4)

Thanks!


On Mon, May 4, 2015 at 5:17 PM, Jeff Alstott <jeffalstott at gmail.com> wrote:

> Hello,
>
> I am trying to use DatABEL do run a coxph model. I am not using DatABEL
> for its intended purpose, of running a straightforward model in which every
> column is a SNP that will be put into the model. Instead, I need to do
> something more complex. I would like to do something like this:
>
> Surv(38579, column1) ~ column2 + column3 + column2*column3 +
> strata(column4)
>
> Is this possible in DatABEL? This is actually just setting up the form to
> do a conditional logistic regression, which if set up correctly is
> equivalent to a Cox proportional hazards model. The survival package
> implements this nicely. I just have far too large of a dataset and want
> DatABEL to help me run the regression.
>
>
> survival's clogit:
> https://stat.ethz.ch/R-manual/R-devel/library/survival/html/clogit.html
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/genabel-devel/attachments/20150505/7754eae7/attachment.html>