Search

In this paper, we develop a new and effective approach to nonparametric quantile regression that accommodates ultrahigh-dimensional data arising from spatio-temporal processes. This approach proves advantageous in staving off computational challenges that constitute known hindrances to existing nonparametric quantile regression methods when the number of predictors is much larger than the available sample size. We investigate conditions under which estimation is feasible and of good overall quality and obtain sharp approximations that we employ to devising statistical inference methodology. These include simultaneous confidence intervals and tests of hypotheses, whose asymptotics is borne by a non-trivial functional central limit theorem tailored to martingale differences. Additionally, we provide finite-sample results through various simulations which, accompanied by an illustrative application to real-worldesque data (on electricity demand), offer guarantees on the performance of the proposed methodology., Comment: 33 pages, 2 figures and accompanying supplementary documentation

Different unsupervised models for dimensionality reduction like PCA, LLE, Shannon's mapping, tSNE, UMAP, etc. work on different principles, hence, they are difficult to compare on the same ground. Although they are usually good for visualisation purposes, they can produce spurious patterns that are not present in the original data, losing its trustability (or credibility). On the other hand, information about some response variable (or knowledge of class labels) allows us to do supervised dimensionality reduction such as SIR, SAVE, etc. which work to reduce the data dimension without hampering its ability to explain the particular response at hand. Therefore, the reduced dataset cannot be used to further analyze its relationship with some other kind of responses, i.e., it loses its generalizability. To make a better dimensionality reduction algorithm with a better balance between these two, we shall formally describe the mathematical model used by dimensionality reduction algorithms and provide two indices to measure these intuitive concepts such as trustability and generalizability. Then, we propose a Localized Skeletonization and Dimensionality Reduction (LSDR) algorithm which approximately achieves optimality in both these indices to some extent. The proposed algorithm has been compared with state-of-the-art algorithms such as tSNE and UMAP and is found to be better overall in preserving global structure while retaining useful local information as well. We also propose some of the possible extensions of LSDR which could make this algorithm universally applicable for various types of data similar to tSNE and UMAP., Comment: This is a dissertation report presented for the completion of the degree of Master of Statistics (M. Stat.) of the author from Indian Statistical Institute

Principal component analysis (PCA) is a widely employed statistical tool used primarily for dimensionality reduction. However, it is known to be adversely affected by the presence of outlying observations in the sample, which is quite common. Robust PCA methods using M-estimators have theoretical benefits, but their robustness drop substantially for high dimensional data. On the other end of the spectrum, robust PCA algorithms solving principal component pursuit or similar optimization problems have high breakdown, but lack theoretical richness and demand high computational power compared to the M-estimators. We introduce a novel robust PCA estimator based on the minimum density power divergence estimator. This combines the theoretical strength of the M-estimators and the minimum divergence estimators with a high breakdown guarantee regardless of data dimension. We present a computationally efficient algorithm for this estimate. Our theoretical findings are supported by extensive simulations and comparisons with existing robust PCA methods. We also showcase the proposed algorithm's applicability on two benchmark datasets and a credit card transactions dataset for fraud detection.

The traditional method of computing singular value decomposition (SVD) of a data matrix is based on a least squares principle, thus, is very sensitive to the presence of outliers. Hence the resulting inferences across different applications using the classical SVD are extremely degraded in the presence of data contamination (e.g., video surveillance background modelling tasks, etc.). A robust singular value decomposition method using the minimum density power divergence estimator (rSVDdpd) has been found to provide a satisfactory solution to this problem and works well in applications. For example, it provides a neat solution to the background modelling problem of video surveillance data in the presence of camera tampering. In this paper, we investigate the theoretical properties of the rSVDdpd estimator such as convergence, equivariance and consistency under reasonable assumptions. Since the dimension of the parameters, i.e., the number of singular values and the dimension of singular vectors can grow linearly with the size of the data, the usual M-estimation theory has to be suitably modified with concentration bounds to establish the asymptotic properties. We believe that we have been able to accomplish this satisfactorily in the present work. We also demonstrate the efficiency of rSVDdpd through extensive simulations., Comment: Author Note: This is a non-peer-reviewed version (only the theoretical part) of the manuscript "Robust singular value decomposition with application to video surveillance background modelling" published in the journal "Statistics and Computing". DOI: 10.1007/s11222-024-10493-7 arXiv admin note: substantial text overlap with arXiv:2109.10680

Robust inference based on the minimization of statistical divergences has proved to be a useful alternative to classical techniques based on maximum likelihood and related methods. Basu et al. (1998) introduced the density power divergence (DPD) family as a measure of discrepancy between two probability density functions and used this family for robust estimation of the parameter for independent and identically distributed data. Ghosh et al. (2017) proposed a more general class of divergence measures, namely the S-divergence family and discussed its usefulness in robust parametric estimation through several asymptotic properties and some numerical illustrations. In this paper, we develop the results concerning the asymptotic breakdown point for the minimum S-divergence estimators (in particular the minimum DPD estimator) under general model setups. The primary result of this paper provides lower bounds to the asymptotic breakdown point of these estimators which are independent of the dimension of the data, in turn corroborating their usefulness in robust inference under high dimensional data., Comment: Keywords: Breakdown Point, Minimum S-divergence Estimator, Density Power Divergence, Power Divergence Details: 28 pages, 7 figures

In real life, we frequently come across data sets that involve some independent explanatory variable(s) generating a set of ordinal responses. These ordinal responses may correspond to an underlying continuous latent variable, which is linearly related to the covariate(s), and takes a particular (ordinal) label depending on whether this latent variable takes value in some suitable interval specified by a pair of (unknown) cut-offs. The most efficient way of estimating the unknown parameters (i.e., the regression coefficients and the cut-offs) is the method of maximum likelihood (ML). However, contamination in the data set either in the form of misspecification of ordinal responses, or the unboundedness of the covariate(s), might destabilize the likelihood function to a great extent where the ML based methodology might lead to completely unreliable inferences. In this paper, we explore a minimum distance estimation procedure based on the popular density power divergence (DPD) to yield robust parameter estimates for the ordinal response model. This paper highlights how the resulting estimator, namely the minimum DPD estimator (MDPDE), can be used as a practical robust alternative to the classical procedures based on the ML. We rigorously develop several theoretical properties of this estimator, and provide extensive simulations to substantiate the theory developed.

A basic algorithmic task in automated video surveillance is to separate background and foreground objects. Camera tampering, noisy videos, low frame rate, etc., pose difficulties in solving the problem. A general approach that classifies the tampered frames, and performs subsequent analysis on the remaining frames after discarding the tampered ones, results in loss of information. Several robust methods based on robust principal component analysis (PCA) have been introduced to solve this problem. To date, considerable effort has been expended to develop robust PCA via Principal Component Pursuit (PCP) methods with reduced computational cost and visually appealing foreground detection. However, the convex optimizations used in these algorithms do not scale well to real-world large datasets due to large matrix inversion steps. Also, an integral component of these foreground detection algorithms is singular value decomposition which is nonrobust. In this paper, we present a new video surveillance background modelling algorithm based on a new robust singular value decomposition technique rSVDdpd which takes care of both these issues. We also demonstrate the superiority of our proposed algorithm on a benchmark dataset and a new real-life video surveillance dataset in the presence of camera tampering. Software codes and additional illustrations are made available at the accompanying website rSVDdpd Homepage (https://subroy13.github.io/rsvddpd-home/)

In this paper, we develop an extension of standard epidemiological models, suitable for COVID-19. This extension incorporates the transmission due to pre-symptomatic or asymptomatic carriers of the virus. Furthermore, this model also captures the spread of the disease due to the movement of people to/from different administrative boundaries within a country. The model describes the probabilistic rise in the number of confirmed cases due to the concomitant effects of (incipient) human transmission and multiple compartments. The associated parameters in the model can help architect the public health policy and operational management of the pandemic. For instance, this model demonstrates that increasing the testing for symptomatic patients does not have any major effect on the progression of the pandemic, but testing rate of the asymptomatic population has an extremely crucial role to play. The model is executed using the data obtained for the state of Chhattisgarh in the Republic of India. The model is shown to have significantly better predictive capability than the other epidemiological models. This model can be readily applied to any administrative boundary (state or country). Moreover, this model can be applied for any other epidemic as well.

Changepoint detection is the problem of finding abrupt or gradual changes in time series data when the distribution of the time series changes significantly. There are many sophisticated statistical algorithms for solving changepoint detection problem, although there is not much work devoted towards gradual changepoints as compared to abrupt ones. Here we present a new approach to solve changepoint detection problem using fuzzy rough set theory which is able to detect such gradual changepoints. An expression for the rough-fuzzy estimate of changepoints is derived along with its mathematical properties concerning fast computation. In a statistical hypothesis testing framework, asymptotic distribution of the proposed statistic on both single and multiple changepoints is derived under null hypothesis enabling multiple changepoint detection. Extensive simulation studies have been performed to investigate how simple crude statistical measures of disparity can be subjected to improve their efficiency in estimation of gradual changepoints. Also, the said rough-fuzzy estimate is robust to signal-to-noise ratio, high degree of fuzziness in true changepoints and also to hyper parameter values. Simulation studies reveal that the proposed method beats other fuzzy methods and also popular crisp methods like WBS, PELT and BOCD in detecting gradual changepoints. The applicability of the estimate is demonstrated using multiple real-life datasets including Covid-19. We have developed the python package "roufcp" for broader dissemination of the methods., Comment: 25 pages, 9 figures, 2 tables

Tea auctions across India occur as an ascending open auction, conducted online. Before the auction, a sample of the tea lot is sent to potential bidders and a group of tea tasters. The seller's reserve price is a confidential function of the tea taster's valuation, which also possibly acts as a signal to the bidders. In this paper, we work with the dataset from a single tea auction house, J Thomas, of tea dust category, on 49 weeks in the time span of 2018-2019, with the following objectives in mind: $\bullet$ Objective classification of the various categories of tea dust (25) into a more manageable, and robust classification of the tea dust, based on source and grades. $\bullet$ Predict which tea lots would be sold in the auction market, and a model for the final price conditioned on sale. $\bullet$ To study the distribution of price and ratio of the sold tea auction lots. $\bullet$ Make a detailed analysis of the information obtained from the tea taster's valuation and its impact on the final auction price. The model used has shown various promising results on cross-validation. The importance of valuation is firmly established through analysis of causal relationship between the valuation and the actual price. The authors hope that this study of the properties and the detailed analysis of the role played by the various factors, would be significant in the decision making process for the players of the auction game, pave the way to remove the manual interference in an attempt to automate the auction procedure, and improve tea quality in markets.

Query by Humming (QBH) is a system to provide a user with the song(s) which the user hums to the system. Current QBH method requires the extraction of onset and pitch information in order to track similarity with various versions of different songs. However, we here focus on detecting precise onsets only and use them to build a QBH system which is better than existing methods in terms of speed and memory and empirically in terms of accuracy. We also provide statistical analogy for onset detection functions and provide a measure of error in our algorithm., Comment: 30 pages, 26 figures