Recent advancement of medical science shows the necessity of customized treatment of complex diseases. Instead of using the same medicine for all patients, this approach recommends targeted therapy for individuals, commonly based on their various biomarkers like genomic profile, demographic information etc. This leads to the identification of the correct subgroup of patients. In this article, we introduce a subgroup identification technique for clustered data using model based partitioning. We develop a semiparametric model based partitioning approach called Quadratic Inference Function Tree (QUIFT). We investigate the performance of this distribution-free method on synthetic data as well as data from a real-world experiment.
Co-author (s): Prof. Kalyan Das (Faculty, IIT Bombay)
Journal: Japanese Journal of Statistics and Data Science
Click here
