Please use this identifier to cite or link to this item: http://hdl.handle.net/2381/39488
Title: Strcs: A command for fitting flexible parametric survival models on the log-hazard scale
Authors: Bower, Hannah
Crowther, Michael J.
Lambert, Paul C.
First Published: 1-Jan-2016
Publisher: StataCorp
Citation: Stata Journal, 2016, 16 (4), pp. 989-1012
Abstract: In this article, we describe strcs, a user-written command for fitting flexible parametric survival models on the log-hazard scale. strcs is an extension of the user-written stgenreg command (Crowther and Lambert, 2013b, Journal of Statistical Software 53(12): 1-17), which fits general parametric models with user-defined hazard functions using numerical integration. strcs implements a two-step method that incorporates both analytical and numerical integration to estimate the cumulative hazard function required for the log-likelihood function. This method improves the accuracy of the fully numeric estimation implemented in stgenreg. Time-dependent effects can be incorporated, and excess mortality models can be fit by using the available options. We also describe some of the extensive postestimation commands that are easily implemented after fitting an strcs model.
ISSN: 1536-867X
eISSN: 1536-8734
Links: http://www.stata-journal.com/article.html?article=st0462
http://hdl.handle.net/2381/39488
Embargo on file until: 1-Jan-10000
Version: Post-print
Status: Peer-reviewed
Type: Journal Article
Rights: Copyright © 2016, StataCorp. Deposited with reference to the publisher’s open access archiving policy.
Description: The file associated with this record is under embargo while permission to archive is sought from the publisher. The full text may be available through the publisher links provided above.
Appears in Collections:Published Articles, Dept. of Health Sciences

Files in This Item:
File Description SizeFormat 
strcs.pdfPost-review (final submitted author manuscript)1.06 MBAdobe PDFView/Open


Items in LRA are protected by copyright, with all rights reserved, unless otherwise indicated.