Out-of-Sample Fusion in Risk Prediction

被引：0
作者：
Myron Katzoff
Wen Zhou
Diba Khan
Guanhua Lu
Benjamin Kedem
机构：
[1] CDC/National Center for Health Statistics,Department of Mathematics
[2] University of Maryland,undefined
来源：
Journal of Statistical Theory and Practice | 2014年 / 8卷 / 3期
关键词：
Mortality; Density ratio model; Threshold probabilities; Tilt; Semiparametric; Coverage; Primary: 62F40; Secondary: 62F25;
D O I：
10.1080/15598608.2013.806233
中图分类号：
学科分类号：
摘要：
The probability that mortality from certain causes exceeds high thresholds is addressed. An out-of-sample fusion method is presented where an original real data sample is fused or combined with independent computer-generated samples in the estimation of exceedance probabilities assuming a density ratio model. Since the size of the combined sample of real and artificial data is larger than that of the real sample, the fused sample produces short confidence intervals relative to traditional methods. Numerical results show that the method maintains good coverage even for some misspecified cases.
引用
收藏
页码：444 / 459
页数：15
相关论文
共 45 条
  • [1] Agresti A(1998)Approximate is better than “exact” for interval estimation of binomial proportions Am. Stat. 52 119-126
  • [2] Coull B A(2001)Interval estimation for a binomial proportion Stat. Sci. 16 101-133
  • [3] Brown L D(2004)Semiparametric density estimation under a two-sample density ratio model Bernoulli 10 583-604
  • [4] Cai T T(1996)Using specially designed exponential families for density estimation Ann. Stat. 24 2431-2461
  • [5] DasGupta A(2004)Merging information for semiparametric density estimation J. R. Stat. Soc. Ser. B 66 941-958
  • [6] Cheng K F(2006)On the effect of misspecifying the density ratio model Ann. Inst. Stat. Math. 58 475-497
  • [7] Chu C K(2001)A semiparametric approach to the one-way layout Technometrics 43 56-65
  • [8] Efron B(2008)A note on Monte Carlo maximization by the density ratio model J. Stat. Theory Pract. 2 355-367
  • [9] Tibshirani R(2000)Large sample theory of maximum likelihood estimates in semiparametric biased sampling models Ann. Stat. 28 151-194
  • [10] Fokianos K(1999)Maximum likelihood estimation in semiparametric selection bias models with application to AIDS vaccine trials Biometrika 86 27-43
12345