Sumitra Purkayastha
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sumitra Purkayastha has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Ayanendranath Basu
Interdisciplinary Statistical Research Unit Indian Statistical Institute Kolkata India
1 shared publication - Moon-Ho Ringo Ho
Nanyang Technological University
1 shared publication - Smarajit Bose
Interdisciplinary Statistical Research Unit Indian Statistical Institute 203 B. T. Road 700108 Kolkata India
1 shared publication - Sourabh Bhattacharya
Institute of Statistics and Decison Sciences Duke University, U.S.A
1 shared publication
Latest Publications
- Open Access
Some incomplete and boundedly complete families of discrete distributions
We present a general result giving us families of incomplete and boundedly complete families of discrete distributions. For such families, the classes of unbiased estimators of zero with finite variance and of parametric functions which will have uniformly minimum variance unbiased estimators with finite variance are explicitly characterized. The general result allows us to construct a large number of families of incomplete and boundedly complete families of discrete distributions. Several new examples of such families are described.
Open AccessSome incomplete and boundedly complete families of discrete distributions
We present a general result giving us families of incomplete and boundedly complete families of discrete distributions. For such families, the classes of unbiased estimators of zero with finite variance and of parametric functions which will have uniformly minimum variance unbiased estimators with finite variance are explicitly characterized. The general result allows us to construct a large number of families of incomplete and boundedly complete families of discrete distributions. Several new examples of such families are described.
- 20 December 2005
- Published by Elsevier in NeuroImage
- Vol. 30 (3) , 794-812
A Bayesian approach to modeling dynamic effective connectivity with fMRI data
- 20 December 2005
- Published by Elsevier in NeuroImage
- Vol. 30 (3) , 794-812
A Bayesian approach to modeling dynamic effective connectivity with fMRI data
- 1 June 2004
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 74 (6) , 445-460
Robust discriminant analysis using weighted likelihood estimators
The procedures in traditional discriminant analysis suffer from serious lack of robustness under model misspecifications. Weighted likelihood estimators based on certain minimum divergence criteria have recently been shown (Markatou et al., 1998 Markatou, M. , Basu, A. and Lindsay, B. G. (1998). Weighted likelihood equations with bootstrap root search. J. Amer. Statist. Assoc., 93: 740–750. [Taylor & Francis Online], [Web of Science ®] [Google Scholar] ) to retain first order efficiency under the model while having attractive robustness properties away from it. In this paper, these estimators have been used to develop classifiers which are robust alternatives to Fisher's discriminant analysis. Results of an extensive simulation study and some real data sets are presented to illustrate the usefulness of the proposed methods.
- 1 June 2004
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 74 (6) , 445-460
Robust discriminant analysis using weighted likelihood estimators
The procedures in traditional discriminant analysis suffer from serious lack of robustness under model misspecifications. Weighted likelihood estimators based on certain minimum divergence criteria have recently been shown (Markatou et al., 1998 Markatou, M. , Basu, A. and Lindsay, B. G. (1998). Weighted likelihood equations with bootstrap root search. J. Amer. Statist. Assoc., 93: 740–750. [Taylor & Francis Online], [Web of Science ®] [Google Scholar] ) to retain first order efficiency under the model while having attractive robustness properties away from it. In this paper, these estimators have been used to develop classifiers which are robust alternatives to Fisher's discriminant analysis. Results of an extensive simulation study and some real data sets are presented to illustrate the usefulness of the proposed methods.