@article{fan_song_lu_2017, title={Change-Plane Analysis for Subgroup Detection and Sample Size Calculation}, volume={112}, ISSN={0162-1459 1537-274X}, url={http://dx.doi.org/10.1080/01621459.2016.1166115}, DOI={10.1080/01621459.2016.1166115}, abstractNote={ABSTRACT We propose a systematic method for testing and identifying a subgroup with an enhanced treatment effect. We adopts a change-plane technique to first test the existence of a subgroup, and then identify the subgroup if the null hypothesis on nonexistence of such a subgroup is rejected. A semiparametric model is considered for the response with an unspecified baseline function and an interaction between a subgroup indicator and treatment. A doubly robust test statistic is constructed based on this model, and asymptotic distributions of the test statistic under both null and local alternative hypotheses are derived. Moreover, a sample size calculation method for subgroup detection is developed based on the proposed statistic. The finite sample performance of the proposed test is evaluated via simulations. Finally, the proposed methods for subgroup identification and sample size calculation are applied to a data from an AIDS study.}, number={518}, journal={Journal of the American Statistical Association}, publisher={Informa UK Limited}, author={Fan, Ailin and Song, Rui and Lu, Wenbin}, year={2017}, month={Apr}, pages={769–778} }
 @article{lewis_fan_2016, title={Improved Acceptance Limits for ASTM Standard E2810}, volume={8}, ISSN={["1946-6315"]}, DOI={10.1080/19466315.2015.1093959}, abstractNote={ASTM Standard E2810 provides a methodology for establishing confidence in passing the USP <905> Uniformity of Dosage Units (UDU) test, and provides acceptance limits for sample means and standard deviations that can be used as elements of lot release. These acceptance limits are quite conservative, however, due to both the nature and shape of an inverted triangular joint confidence region for the lot mean and standard deviation. We obtain improved (wider) acceptance limits by using a Bayesian approach that focuses on the posterior distribution of the probability of passing the USP <905> UDU test. The Bayesian approach has good sampling properties, and the improvement in acceptance limits can be considerable. For example, for sample size of 10 units and sample means between 97 and 103, the Bayesian approach results in acceptance limits for sample standard deviations that are at least 25% greater than those in E2810. The impact of the improved acceptance limits is illustrated with operating characteristic curves.}, number={1}, journal={STATISTICS IN BIOPHARMACEUTICAL RESEARCH}, author={Lewis, Richard A. and Fan, Ailin}, year={2016}, month={Jan}, pages={40–48} }
 @article{fan_lu_song_2016, title={SEQUENTIAL ADVANTAGE SELECTION FOR OPTIMAL TREATMENT REGIME}, volume={10}, ISSN={["1932-6157"]}, DOI={10.1214/15-aoas849}, abstractNote={Variable selection for optimal treatment regime in a clinical trial or an observational study is getting more attention. Most existing variable selection techniques focused on selecting variables that are important for prediction, therefore some variables that are poor in prediction but are critical for decision-making may be ignored. A qualitative interaction of a variable with treatment arises when treatment effect changes direction as the value of this variable varies. The qualitative interaction indicates the importance of this variable for decision-making. Gunter, Zhu and Murphy (2011) proposed S-score which characterizes the magnitude of qualitative interaction of each variable with treatment individually. In this article, we developed a sequential advantage selection method based on the modified S-score. Our method selects qualitatively interacted variables sequentially, and hence excludes marginally important but jointly unimportant variables or vice versa. The optimal treatment regime based on variables selected via joint model is more comprehensive and reliable. With the proposed stopping criteria, our method can handle a large amount of covariates even if sample size is small. Simulation results show our method performs well in practical settings. We further applied our method to data from a clinical trial for depression.}, number={1}, journal={ANNALS OF APPLIED STATISTICS}, author={Fan, Ailin and Lu, Wenbin and Song, Rui}, year={2016}, month={Mar}, pages={32–53} }