Translator Disclaimer
February 2020 Consistent semiparametric estimators for recurrent event times models with application to virtual age models
Eric Beutner, Laurent Bordes, Laurent Doyen
Bernoulli 26(1): 557-586 (February 2020). DOI: 10.3150/19-BEJ1140
Supplemental Content

Supplement A.1: Additional proofs. This supplement material gives the proof of Lemmas 3.1 and 3.3, details the proof of Assumptions B2 and D1, and also demonstrates that Assumptions B1–B3 and D1–D2 are satisfied under alternative censoring schemes.

Digital Object Identifier: 10.3150/19-BEJ1140SUPPA

Supplement A.2: Data set. This supplement material displays the data set considered in Section 5.

Digital Object Identifier: 10.3150/19-BEJ1140SUPPB

Supplement B.1: Numercial illustration. File “VStheta_CensI.mp4” represents the empirical bias, standard deviation and Mean Square Error (MSE) of $N=1000$ estimates of $\theta_{0}$ versus the true parameter value, for an $\mathrm{ARA}_{1}$ model (red) and an $\mathrm{ARA}_{\infty }$ model (blue) with sample size $n=50$ (dotted lines), $n=100$ (dashed dotted lines), $n=200$ (dashed lines), $n=400$ (long dashed lines) for Type-I censoring with $\tau =s=5$ and bandwidth $b$ varying in $[0.1,1]$.

Digital Object Identifier: 10.3150/19-BEJ1140SUPPC

Supplement B.2: Numercial illustration. File “VStheta_CensII.mp4” is similar to Supplement B.1, but for Type-II censoring with $k=4$ ($s$ sufficiently large).

Digital Object Identifier: 10.3150/19-BEJ1140SUPPD

Supplement B.3: Numercial illustration. File “VSb_ CensI.mp4” is similar to Supplement B.1, but plots are versus bandwidth $b$, for Type-I censoring with $\tau =s=5$ and a true parameter value $\theta_{0}$ varying in $[0,1]$.

Digital Object Identifier: 10.3150/19-BEJ1140SUPPE

Supplement B.4: Numercial illustration. File “VSb_ CensII.mp4” is similar to Supplement B.3, but for Type-II censoring with $k=4$ ($s$ sufficiently large).

Digital Object Identifier: 10.3150/19-BEJ1140SUPPF

Supplement B.5: Numercial illustration. File “VSs_CensII.mp4” is similar to Supplement B.1, but plots are versus $s$, for Type-II censoring with $k=4$ ($s$ sufficiently large) and a true parameter value $\theta_{0}$ varying in $[0,1]$ and $b=0.5$.

Digital Object Identifier: 10.3150/19-BEJ1140SUPPG
JOURNAL ARTICLE
30 PAGES

This article is only available to subscribers.
It is not available for individual sale.
+ SAVE TO MY LIBRARY

SHARE
Vol.26 • No. 1 • February 2020
Back to Top