bootstrap for cox proportional hazards model

[TomDoan]

Have you checked the literature? Offhand, that doesn't look right.

[fan]

Have you checked the literature? Offhand, that doesn't look right.

The literature I could find on this matter is largely about independent bootstrapping in samples, the only paper I find could be related to what I want is the paper written by Thomas Loughin (2007), "A residual bootstrap for regression parameters in proportional hazards models" . But I do not have the access to the paper. If it is possible, could you please kindly share the paper with me if you are able to have the paper? Many Thanks.

[TomDoan]

The first hit on get on Googling "bootstrap proportional hazard" is his dissertation:

http://lib.dr.iastate.edu/cgi/viewconte ... ontext=rtd

[fan]

The first hit on get on Googling "bootstrap proportional hazard" is his dissertation:

http://lib.dr.iastate.edu/cgi/viewconte ... ontext=rtd

Thank you for the link. I think I made a mistake in my initial data set. The author says

"The focus of this study is a set of covariates representing specified periods before or after elections. These preelection and postelection periods are identified by dummy variables that enter as time-varying covariates, that is, covariates that can change over the course of a spell.8 Each of these covariates represents one of three different time frames and correspondingly is set equal to 1 in the nine-month, twelve-month or twenty-four-month period either before or after an election. The use of different time frames for the covariates allows for investigation of the length of the period of the political effect on the hazard. The coefficients across different specifications of time frames are directly comparable since the hazard is the estimate of the likelihood of the completion of a business cycle phase in the next month conditional on its lasting up until that month."

Since the covariates are time-dependent, so I am wondering what are instruments for Cox ph model with time-dependent covairates. It is my first time to use time-dependent covariates in a study, could you please kindly take a quick look at my data sample to see whether it is in the right format? Thank you in advance

[TomDoan]

The question of whether you are properly interpreting that paper strikes me as something that you discuss with your supervisor---it has nothing to do with the software. Also, if you are trying to educate yourself on proportional hazard models, you should probably seek out a biometrician at your university. Hazard models are heavily used in biometrics, but rarely are used in econometrics.