In recent days, a combination of finite mixture model (FMM) and hidden Markov model (HMM) is becoming popular for partitioning heterogeneous temporal data into homogeneous groups (clusters) with homogeneous time points (regimes). The regression mixtures commonly considered in this approach can also accommodate for covariates present in data. The classical fixed covariate approach, however, may not always serve as a reasonable assumption as it is incapable of accounting for the contribution of covariates in cluster formation. This paper introduces a novel approach for detecting clusters and regimes in time series data in the presence of random covariates. The computational challenges related to the proposed model has been discussed, and several simulation studies are performed. An application to United States COVID‐19 data yields meaningful clusters and regimes.
Optical absorption in semiconductors is one of the most basic yet complex properties in crystalline semiconductors. There have been decades of studies on optical absorption with specific efforts towards understanding the accuracy of measurements. However, the interdependence of the parameters employed in the fit routines poses challenges in calculating consistent and reliable errors associated with the process when the band tailing states are involved, which is unavoidable in all practical measurements. This work examines room-temperature absorption measurements in undoped GaAs to report consistent error bars employing a bootstrap resampling analysis. The results show that (1) the transmissive part of the measured absorption contains all the information about the fit parameters for the entire spectrum and (2) it is possible to utilize statistical bootstrap sampling to determine error bars when interdependent fit parameters are present in the fit model.
3 August 2024 in Data Mining and Knowledge Discovery
Finding clusters in a dynamic trade network and identifying anomalies in the trading pattern is a problem of great interest. In this paper, a novel method of clustering dynamic multilayer weighted network relying on a mixture of tensor normal distribution has been proposed. Upon finding the clusters of nodes exhibiting similar trading relationships over the years, atypical trading exchanges have been identified that deviate from the regular pattern. The developed methodology has been applied to a European trade network consisting of 39 countries observed over a decade.