R help - Mar 2009 - logistic regression model validation through bootstrapping

Hi,

I was wondering whether this query was addressed on how to perform
validation through boostrapping. I am currently trying to implement a
boostrapping approach to validation but don't know where to start. Help
please.


Thank you and Regards,

Vivienne Ozohili
Risk Model Validation Manager
Group Risk
Independent Control Unit
A5, Bank of Ireland Head Office
Lower Baggot Street
Dublin 2
Tel:  +353-01-6044833
Email: Vivienne_O.Ozohili@boimail.com

The Governor and Company of the Bank of Ireland is regulated by the
Financial Regulator in Ireland and authorised by the Financial Services
Authority in the UK. Bank of Ireland incorporated in Ireland with Limited
Liability.
Registered Office: Head Office, Lower Baggot Street, Dublin 2. Registered
Number - C-1
<PRE>
**************************
</PRE>
This email and any files transmitted with it are confide...{{dropped:15}}

Vivienne Ozohili wrote:
>> I am currently trying to implement a boostrapping approach to
validation
>> but don't know where to start. Help please.
Take a look at Prof. Harrell's ?validate.lrm in package Design.

Regards, Mark.


Vivienne_O.Ozohili wrote:
> 
> 
> Hi,
> 
> I was wondering whether this query was addressed on how to perform
> validation through boostrapping. I am currently trying to implement a
> boostrapping approach to validation but don't know where to start. Help
> please.
> 
> 
> Thank you and Regards,
> 
> Vivienne Ozohili
> Risk Model Validation Manager
> Group Risk
> Independent Control Unit
> A5, Bank of Ireland Head Office
> Lower Baggot Street
> Dublin 2
> Tel:  +353-01-6044833
> Email: Vivienne_O.Ozohili at boimail.com
> 
> The Governor and Company of the Bank of Ireland is regulated by the
> Financial Regulator in Ireland and authorised by the Financial Services
> Authority in the UK. Bank of Ireland incorporated in Ireland with Limited
> Liability.
> Registered Office: Head Office, Lower Baggot Street, Dublin 2. Registered
> Number - C-1
> <PRE>
> **************************
> </PRE>
> This email and any files transmitted with it are confide...{{dropped:15}}
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
> 
-- 
View this message in context:
http://www.nabble.com/Re%3A-logistic-regression-model-validation-through-bootstrapping-tp22293497p22293820.html
Sent from the R help mailing list archive at Nabble.com.

Install the Design and Hmisc package then do:
library(Design)
?validate.lrm
?calibrate

Resampling methods such as the bootstrap and cross-validation assume you 
have done no model/variable selection outside the bootstrap look, i.e., 
you let the bootstrap repeat all modeling steps that used Y.  validate 
and calibrate only support backward stepdown variable selection, but if 
you are using full pre-specified model fits, this is even easier.

Frank


Vivienne_O.Ozohili at boimail.com wrote:
> Hi,
> 
> I was wondering whether this query was addressed on how to perform
> validation through boostrapping. I am currently trying to implement a
> boostrapping approach to validation but don't know where to start. Help
> please.
> 
> 
> Thank you and Regards,
> 
> Vivienne Ozohili
> Risk Model Validation Manager
> Group Risk
> Independent Control Unit
> A5, Bank of Ireland Head Office
> Lower Baggot Street
> Dublin 2
> Tel:  +353-01-6044833
> Email: Vivienne_O.Ozohili at boimail.com
> 
> The Governor and Company of the Bank of Ireland is regulated by the
> Financial Regulator in Ireland and authorised by the Financial Services
> Authority in the UK. Bank of Ireland incorporated in Ireland with Limited
> Liability.
> Registered Office: Head Office, Lower Baggot Street, Dublin 2. Registered
> Number - C-1
> <PRE>
> **************************
> </PRE>
> This email and any files transmitted with it are confide...{{dropped:15}}
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 

-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University