不完全无误判金标准下二重抽样设计中样本量的确定
|
摘要
考虑在不完全无误判金标准下二重抽样设计中对疾病流行率进行检验的近似样本量公式。基于两种模型提出了在给定置信水平下置信区间宽度控制在指定范围内的样本量的近似公式,随机模拟研究了在估计的样本量下区间估计的统计性质。结果表明:基于限制性极大似然估计下方差的Wald置信区间、似然比置信区间和Score置信区间确定的样本量是准确有效的,因而被推荐于实际应用中。实际数据分析进一步验证了该方法的有效性。
We consider the approximate sample size formula for testing the disease prevalence in double-sampling design with imperfect gold standard. Sample size formulas are derived to control the width of a confidence interval at a specified width with a pre-specified confidence level under two models. Simulation study is conducted to investigate the performance of various formulas,empirical results show that sample sizes based on Wald confidence interval with constrained maximum likelihood variance,likelihood ratio confidence interval and score confidence interval are accurate and are hence recommended in practical applications. The applicability of the proposed methods is illustrated by a real-data example.
We consider the approximate sample size formula for testing the disease prevalence in double-sampling design with imperfect gold standard. Sample size formulas are derived to control the width of a confidence interval at a specified width with a pre-specified confidence level under two models. Simulation study is conducted to investigate the performance of various formulas,empirical results show that sample sizes based on Wald confidence interval with constrained maximum likelihood variance,likelihood ratio confidence interval and score confidence interval are accurate and are hence recommended in practical applications. The applicability of the proposed methods is illustrated by a real-data example.
引文
[1]TENENBEIN A A.A double sampling scheme for estimating from binomial data with misclassifications[J].Journal of the A-merican Statistical Association,1970,65:1350-1361.
[2]YIU C F,POON W Y.Estimating the polychoric correlation from misclassified data[J].British Journal of Mathematical and Statistical Psychology,2008,61:133-161.
[3]TANG M L,QIU S F,POON W Y,et al.Test procedures for disease prevalence with partially validated data[J].Journal of Biopharmaceutical Statistics,2012,22:368-386.
[4]BOESE D H,YOUNG D M,STAMEY J D.Confidence intervals for a binomial parameter based on binary data subject to falsepositive misclassification[J].Computational Statatistics and Data Analysis,2006,50:3369-3385.
[5]MORVAN J,COSTE J,ROUX C H,et al.Guillemin F.Prevalence in two-phase surveys:accuracy of screening procedure and corrected estimates[J].Annuals of Epidemiology,2008,18:261-269.
[6]TANG M L,QIU S F,POON W Y.Confidence interval construction for disease prevalence based on partial validation series[J].Computational Statistics and Data Analysis,2012,56:1200-1220.
[7]QIU S F,POON W Y,TANG M L.Sample size determination for disease prevalence studies with partially validated data[J].Statistical Methods in Medical Research,2016,25(1):37-63.
[8]TANG M L,QIU S F,POON W Y.Comparison of disease prevalence in two populations in presence of misclassification[J].Biometrical Journal,2012,54(6):786-807.
[9]QIU S F,POON W Y,TANG M L.Confidence intervals for proportion difference from two independent partially validated series[J].Statistical Methods in Medical Research,2016,25(5):2250-2273.
[10]NEDELMAN J.The prevalence of malaria in Garki,Nideria:double sampling with a fallible expert[J].Biometrics,1988,44(3):635-655.
[11]QIU S F,LIAN H,ZOU G Y,et al.Interval estimation for a proportion using a double sampling scheme with two fallible classifiers[J].Statistical Methods in Medical Research,2016.DOI:10.1177/0962280216681599.
[12]LIU R T,HEUCH I,IRGENS L M.Maximum likelihood estimation of the proportion of congenital malformation using double registration systems[J].Biometrics,1994,50:433-444.
[2]YIU C F,POON W Y.Estimating the polychoric correlation from misclassified data[J].British Journal of Mathematical and Statistical Psychology,2008,61:133-161.
[3]TANG M L,QIU S F,POON W Y,et al.Test procedures for disease prevalence with partially validated data[J].Journal of Biopharmaceutical Statistics,2012,22:368-386.
[4]BOESE D H,YOUNG D M,STAMEY J D.Confidence intervals for a binomial parameter based on binary data subject to falsepositive misclassification[J].Computational Statatistics and Data Analysis,2006,50:3369-3385.
[5]MORVAN J,COSTE J,ROUX C H,et al.Guillemin F.Prevalence in two-phase surveys:accuracy of screening procedure and corrected estimates[J].Annuals of Epidemiology,2008,18:261-269.
[6]TANG M L,QIU S F,POON W Y.Confidence interval construction for disease prevalence based on partial validation series[J].Computational Statistics and Data Analysis,2012,56:1200-1220.
[7]QIU S F,POON W Y,TANG M L.Sample size determination for disease prevalence studies with partially validated data[J].Statistical Methods in Medical Research,2016,25(1):37-63.
[8]TANG M L,QIU S F,POON W Y.Comparison of disease prevalence in two populations in presence of misclassification[J].Biometrical Journal,2012,54(6):786-807.
[9]QIU S F,POON W Y,TANG M L.Confidence intervals for proportion difference from two independent partially validated series[J].Statistical Methods in Medical Research,2016,25(5):2250-2273.
[10]NEDELMAN J.The prevalence of malaria in Garki,Nideria:double sampling with a fallible expert[J].Biometrics,1988,44(3):635-655.
[11]QIU S F,LIAN H,ZOU G Y,et al.Interval estimation for a proportion using a double sampling scheme with two fallible classifiers[J].Statistical Methods in Medical Research,2016.DOI:10.1177/0962280216681599.
[12]LIU R T,HEUCH I,IRGENS L M.Maximum likelihood estimation of the proportion of congenital malformation using double registration systems[J].Biometrics,1994,50:433-444.