We introduce a robust likelihood approach to inference about marginal distributional characteristics for paired data without modeling correlation/joint probabilities. This method is reproducible in that it is applicable to paired settings with various sizes. The virtue of the new strategy is elucidated via testing marginal homogeneity in paired triplet scenario. We use simulations and real data analysis to demonstrate the merit of our robust likelihood methodology.
TsouTS. A robust likelihood approach to inference about the difference between two multinomial distributions in paired designs. Stat Meth Med Res2018; 27: 3077–3091.
6.
TsouTS. A new likelihood approach to inference about predictive values of diagnostic tests in paired designs. Stat Meth Med Res2018; 27: 541–548.
7.
MiyashitaNKawaiYYamaguchiT, et al.Clinical potential of diagnostic methods for the rapid diagnosis of Mycoplasma pneumoniae pneumonia in adults. Eur J Clin Microbiol Infect Dis2011; 30: 439–446.
8.
LiuJPHsuehHMHsiehE, et al.Tests for equivalence or non-inferiority for paired binary data. Stat Med2002; 21: 231–245.
9.
KosinskiAS. A weighted generalized score statistic for comparison of predictive values of diagnostic tests. Stat Med2013; 32: 964–977.
10.
RoyallRMTsouTS. Interpreting statistical evidence by using imperfect models: robust adjusted likelihood functions. J Royal Stat Soc Ser B2003; 65: 391–404.
