P
Prafulla Chandra
Researcher at Indian Institute of Technology Madras
Publications - 8
Citations - 15
Prafulla Chandra is an academic researcher from Indian Institute of Technology Madras. The author has contributed to research in topics: Estimator & Mean squared error. The author has an hindex of 2, co-authored 7 publications receiving 9 citations.
Papers
More filters
Proceedings ArticleDOI
Concentration and Tail Bounds for Missing Mass
TL;DR: This work improves upon the best-known left and right tail bounds for missing mass and provides a new bounding method for the moment generating function of a generalized version of missing mass that results in a noticeable improvement in the tail bound.
Proceedings ArticleDOI
Improved Tail Bounds for Missing Mass and Confidence Intervals for Good-Turing Estimator
TL;DR: This work shows that missing mass is sub-Gamma on the right tail with the best-possible variance parameter under the Poisson and multinomial sampling models, and derives confidence intervals for the Good-Turing estimator with better confidence levels and narrower width when compared to existing ones.
Proceedings ArticleDOI
Convergence of Chao Unseen Species Estimator
TL;DR: In this article, the authors analyze the Chao estimator and show that its worst case mean squared error (MSE) is smaller than the MSE of the plug-in estimator by a factor of
Proceedings ArticleDOI
Missing Mass of Markov Chains
TL;DR: A useful ‘multi-letter’ characterization of the bias leads to sufficient conditions on the transition probability matrix for convergence of the biases of the Good-Turing estimator for missing mass in Markov chains to zero.
Posted Content
Missing Mass of Rank-2 Markov Chains.
TL;DR: In this paper, an upper bound on the absolute bias of the Good-Turing (GT) estimator in terms of the spectral gap of the chain and the occupancy of states is derived for rank-2 irreducible Markov chains.