Imperial College London
Portrait Picture

DrRogerNewson

Faculty of Medicine, School of Public Health

Honorary Research Associate
 
 
 
//

Contact

 

+44 (0)20 7594 2784r.newson Website

 
 
//

Assistant

 

Ms Dorothea Cockerell +44 (0)20 7594 3368

 
//

Location

 

351Reynolds BuildingCharing Cross Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Newson:2003,
author = {Newson, R and Team, TALSPACS},
journal = {Stata Journal},
pages = {109--132},
title = {Multiple-test procedures and smile plots},
url = {http://hdl.handle.net/10044/1/27162},
volume = {3},
year = {2003}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - multproc carries out multiple-test procedures, taking as input a list of p-valuesand an uncorrected critical p-value, and calculating a corrected overall critical pvaluefor rejection of null hypotheses. These procedures define a conÞdence regionfor a set-valued parameter, namely the set of null hypotheses that are true. Theyaim to control either the family-wise error rate (FWER) or the false discoveryrate (FDR) at a level no greater than the uncorrected critical p-value. smileplotcalls multproc and then creates a smile plot, with data points corresponding toestimated parameters, the p-values (on a reverse log scale) on the y-axis, and theparameter estimates (or another variable) on the x-axis. There are y-axis referencelines at the uncorrected and corrected overall critical p-values. The reference linefor the corrected overall critical p-value, known as the parapet line, is an informalÒupper confidence limitÓ for the set of null hypotheses that are true and defines aboundary between data mining and data dredging. A smile plot summarizes a setof multiple analyses just as a Cochrane forest plot summarizes a meta-analysis.   Copyright 2003 by Stata Corporation.
AU  - Newson,R
AU  - Team,TALSPACS
EP  - 132
PY  - 2003///
SP  - 109
TI  - Multiple-test procedures and smile plots
T2  - Stata Journal
UR  - http://hdl.handle.net/10044/1/27162
VL  - 3
ER  -
Download