Re: st:spline/fractional polynomials with survey data

[Roger Newson <r.newson@imperial.ac.uk>]

There is no reason why users should not use spline models with 
probability weights, sample survey strata, clustering, or even finite 
population corrections, with or without bootstrapping or jackknifing. 
However, they should probably not expect a one-line solution. Instead, 
they should produce the spline basis with one command, and then fit the 
regression model (with the spline basis included in the design matrix) 
using a regression command, with the -svy:- prefix.

My own package for creating a spline basis is the -bspline- package, 
which you can download from SSC. There are 3 modules in that package, 
but the one to use is probably -flexcurv-. The -flexcurv- module inputs 
an X-variable, in which the spline is to be defined, and outputs a basis 
(or incomplete basis) of reference splines, for inclusion in a design 
matrix. If these reference splines are included in a design matrix, then 
the corresponding parameters will be values of the spline at reference 
points on the X-axis, or possibly differences between values of the 
spline at reference points and values of the spline at a baseline 
reference point. These parameters can be understood by non-technical 
people (I think), because they play the role for continuous factors that 
dummy variable parameters (as produced using -xi:-) play for discrete 
factors. The -bspline- package is distributed with a .pdf manual, which 
illustrates the use of the package with examples.

I hope this helps.

Best wishes

Roger


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: r.newson@imperial.ac.uk
Web page: http://www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/popgenetics/reph/

Opinions expressed are those of the author, not of the institution.

On 01/04/2011 11:59, Ronald McDowell wrote:
> Has anyone here used spline methods or fractional polynomial methods in conjunction with survey data, or come across some references in that area?
>
> I see both commands allow for weighting, but I have both (probability) weights and stratification to content with.
>
> Many thanks for any feedback.
>
> Ron
>
> -----------------------------------------
> Ron McDowell
> Institute of Nursing Research
> University of Ulster, Coleraine
> McDowell-R3@email.ulster.ac.uk
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/