×

Nonparametric estimation of conditional transition probabilities in a non-Markov illness-death model. (English) Zbl 1317.65041

Summary: One important goal in multi-state modeling is the estimation of transition probabilities. In longitudinal medical studies these quantities are particularly of interest since they allow for long-term predictions of the process. In recent years significant contributions have been made regarding this topic. However, most of the approaches assume independent censoring and do not account for the influence of covariates. The goal of the paper is to introduce feasible estimation methods for the transition probabilities in an illness-death model conditionally on current or past covariate measures. All approaches are evaluated through a simulation study, leading to a comparison of two different estimators. The proposed methods are illustrated using a real colon cancer data set.

MSC:

62-08 Computational methods for problems pertaining to statistics
62G05 Nonparametric estimation
62P10 Applications of statistics to biology and medical sciences; meta analysis
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] Aalen O, Johansen S (1978) An empirical transition matrix for non homogeneous markov and chains based on censored observations. Scand J Stat 5:141-150 · Zbl 0383.62058
[2] Akritas M (1994) Nearest neighbor estimation of a bivariate distribution under random censoring. Ann Stat 22:1299-1327 · Zbl 0819.62028 · doi:10.1214/aos/1176325630
[3] Altman N, Leger C (1995) Bandwidth selection for kernel distribution function estimation. J Stat Plan Inference 46:195-214 · Zbl 0833.62035 · doi:10.1016/0378-3758(94)00102-2
[4] Amorim A, de Uña-Álvarez J, Meira-Machado L (2011) Presmoothing the transition probabilities in the illness-death model. Stat Probab Lett 81(7):797-806 · Zbl 1218.62111 · doi:10.1016/j.spl.2011.02.017
[5] Andersen P, Borgan Ø, Gill R, Keiding N (1993) Statistical models based on counting processes. Springer, New York · Zbl 0769.62061 · doi:10.1007/978-1-4612-4348-9
[6] Arjas E, Eerola M (1993) On predictive causality in longitudinal studies. J Stat Plan Inference 34:361-384 · Zbl 0766.60048 · doi:10.1016/0378-3758(93)90146-W
[7] Beran R (1981) Nonparametric regression with randomly censored survival data. Technical report. University of California, Berkeley
[8] Breslow NE (1972) Discussion of the paper by D. R. Cox. J R Stat Soc Ser B 34:216-217
[9] Cox DR (1972) Regression model and life-tables. J R Stat Soc Ser B 34(2):187-220 · Zbl 0243.62041
[10] Dabrowska D (1987) Non-parametric regression with censored survival data. Scand J Stat 14:181-197 · Zbl 0641.62024
[11] Dabrowska D, Ho W (2000) Confidence bands for comparison of transition probabilities in a markov chain model. Lifetime Data Anal 6(1):5-21 · Zbl 0948.62064 · doi:10.1023/A:1009681332533
[12] Dabrowska D, Lee W (1996) Nonparametric estimation of transition probabilities in a two-stage duration model. J Nonparametr Stat 7:75-103 · Zbl 0872.62037 · doi:10.1080/10485259608832692
[13] Kaplan E, Meier P (1958) Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481 · Zbl 0089.14801 · doi:10.1080/01621459.1958.10501452
[14] Kyung-Joon C, Schucany W (1998) Nonparametric kernel regression estimation near endpoints. J Stat Plan Inference 66:289-304 · Zbl 1041.62507 · doi:10.1016/S0378-3758(97)00082-7
[15] Lin D, Sun W, Ying Z (1999) Nonparametric estimation of the gap time distributions for serial events with censored data. Biometrika 86:59-70 · Zbl 0917.62030 · doi:10.1093/biomet/86.1.59
[16] Meira-Machado L, Roca-Pardiñas J (2011) p3state.msm: analyzing survival data from an illness-death model. J Stat Softw 38:1-18
[17] Meira-Machado L, de Uña-Álvarez J, Cadarso-Suárez C (2006) Nonparametric estimation of transition probabilities in a non-markov illness-death model. Lifetime Data Anal 12:325-344 · Zbl 1356.62127 · doi:10.1007/s10985-006-9009-x
[18] Meira-Machado L, deUña-Álvarez J, Cadarso-Suárez C, Andersen P (2009) Multi-state models for the analysis of time to event data. Stat Methods Med Res 18:195-222 · doi:10.1177/0962280208092301
[19] Moertel C, Fleming T, McDonald ea JS (1990) Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 322:352-358 · doi:10.1056/NEJM199002083220602
[20] Mostajabi F, Datta S (2013) Nonparametric regression of state occupation, entry, exit and waiting times with multistate right censored data. Stat Med 32:3006-3019 · doi:10.1002/sim.5703
[21] Nadaraya E (1965) On nonparametric estimates of density functions and regression curves. Theory Appl Probab 10:186-190 · Zbl 0134.36302 · doi:10.1137/1110024
[22] Satten G, Datta S (2001) The Kaplan-Meier estimator as an inverse-probability-of-censoring weighted average. Am Stat 55(3):207-210 · Zbl 1182.62191 · doi:10.1198/000313001317098185
[23] Strzalkowska-Kominiak E, Cao R (2013) Maximum likelihood estimation for conditional distribution single-index models under censoring. J Multivar Anal 114:74-96 · Zbl 1258.62045
[24] Strzalkowska-Kominiak E, Cao R (2014) Beran-based approach for single-index models under censoring. Comput Stat 29(5):1243-1261 · Zbl 1306.65139
[25] Van Keilegom I (2004) A note on the nonparametric estimation of the bivariate distribution under dependent censoring. J Nonparametr Stat 16:659-670 · Zbl 1074.62026 · doi:10.1080/10485250310001624783
[26] Van Keilegom I, Akritas M, Veraverbeke N (2001) Estimation of the conditional distribution in regression with censored data: a comparative study. Comput Stat Data Anal 35:487-500 · Zbl 1080.62515 · doi:10.1016/S0167-9473(00)00025-6
[27] Wand M, Jones M (1997) Kernel smoothing. Chapman & Hall, London · Zbl 0854.62043
[28] Watson G (1964) Smooth regression analysis. Sankhya 26(15):175-184 · Zbl 0128.13503
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.