-
A New Spatio-Temporal Model Exploiting Hamiltonian Equations
Authors:
Satyaki Mazumder,
Sayantan Banerjee,
Sourabh Bhattacharya
Abstract:
The solutions of Hamiltonian equations are known to describe the underlying phase space of a mechanical system. In this article, we propose a novel spatio-temporal model using a strategic modification of the Hamiltonian equations, incorporating appropriate stochasticity via Gaussian processes. The resultant spatio-temporal process, continuously varying with time, turns out to be nonparametric, non…
▽ More
The solutions of Hamiltonian equations are known to describe the underlying phase space of a mechanical system. In this article, we propose a novel spatio-temporal model using a strategic modification of the Hamiltonian equations, incorporating appropriate stochasticity via Gaussian processes. The resultant spatio-temporal process, continuously varying with time, turns out to be nonparametric, non-stationary, non-separable, and non-Gaussian. Additionally, the lagged correlations converge to zero as the spatio-temporal lag goes to infinity. We investigate the theoretical properties of the new spatio-temporal process, including its continuity and smoothness properties. We derive methods for complete Bayesian inference using MCMC techniques in the Bayesian paradigm. The performance of our method has been compared with that of a non-stationary Gaussian process (GP) using two simulation studies, where our method shows a significant improvement over the non-stationary GP. Further, applying our new model to two real data sets revealed encouraging performance.
△ Less
Submitted 7 December, 2024; v1 submitted 24 November, 2022;
originally announced November 2022.
-
Functional Classwise Principal Component Analysis: A Novel Classification Framework
Authors:
Avishek Chatterjee,
Satyaki Mazumder,
Koel Das
Abstract:
In recent times, functional data analysis (FDA) has been successfully applied in the field of high dimensional data classification. In this paper, we present a novel classification framework using functional data and classwise Principal Component Analysis (PCA). Our proposed method can be used in high dimensional time series data which typically suffers from small sample size problem. Our method e…
▽ More
In recent times, functional data analysis (FDA) has been successfully applied in the field of high dimensional data classification. In this paper, we present a novel classification framework using functional data and classwise Principal Component Analysis (PCA). Our proposed method can be used in high dimensional time series data which typically suffers from small sample size problem. Our method extracts a piece wise linear functional feature space and is particularly suitable for hard classification problems.The proposed framework converts time series data into functional data and uses classwise functional PCA for feature extraction followed by classification using a Bayesian linear classifier. We demonstrate the efficacy of our proposed method by applying it to both synthetic data sets and real time series data from diverse fields including but not limited to neuroscience, food science, medical sciences and chemometrics.
△ Less
Submitted 26 June, 2021;
originally announced June 2021.
-
Nonparametric Dynamic State Space Modeling of Observed Circular Time Series with Circular Latent States: A Bayesian Perspective
Authors:
Satyaki Mazumder,
Sourabh Bhattacharya
Abstract:
Circular time series has received relatively little attention in statistics and modeling complex circular time series using the state space approach is non-existent in the literature. In this article we introduce a flexible Bayesian nonparametric approach to state space modeling of observed circular time series where even the latent states are circular random variables. Crucially, we assume that t…
▽ More
Circular time series has received relatively little attention in statistics and modeling complex circular time series using the state space approach is non-existent in the literature. In this article we introduce a flexible Bayesian nonparametric approach to state space modeling of observed circular time series where even the latent states are circular random variables. Crucially, we assume that the forms of both observational and evolutionary functions, both of which are circular in nature, are unknown and time-varying. We model these unknown circular functions by appropriate wrapped Gaussian processes having desirable properties.
We develop an effective Markov chain Monte Carlo strategy for implementing our Bayesian model, by judiciously combining Gibbs sampling and Metropolis-Hastings methods. Validation of our ideas with a simulation study and two real bivariate circular time series data sets, where we assume one of the variables to be unobserved, revealed very encouraging performance of our model and methods.
We finally analyse a data consisting of directions of whale migration, considering the unobserved ocean current direction as the latent circular process of interest. The results that we obtain are encouraging, and the posterior predictive distribution of the observed process correctly predicts the observed whale movement.
△ Less
Submitted 15 March, 2017; v1 submitted 26 October, 2016;
originally announced October 2016.
-
On existence of a change in mean of functional data
Authors:
Buddhananda Banerjee,
Satyaki Mazumder
Abstract:
Functional data often arise as sequential temporal observations over a continuous state-space. A set of functional data with a possible change in its structure may lead to a wrong conclusion if it is not taken in to account. So, sometimes, it is crucial to know about the existence of change point in a given sequence of functional data before doing any further statistical inference. We develop a ne…
▽ More
Functional data often arise as sequential temporal observations over a continuous state-space. A set of functional data with a possible change in its structure may lead to a wrong conclusion if it is not taken in to account. So, sometimes, it is crucial to know about the existence of change point in a given sequence of functional data before doing any further statistical inference. We develop a new methodology to provide a test for detecting a change in the mean function of the corresponding data. To obtain the test statistic we provide an alternative estimator of the covariance kernel. The proposed estimator is asymptotically unbiased under the null hypothesis and, at the same time, has smaller amount of bias than that of the existing estimator. We show here that under the null hypothesis the proposed test statistic is pivotal asymptotically. Moreover, it is shown that under alternative hypothesis the test is consistent for large enough sample size. It is also found that the proposed test is more powerful than the available test procedure in the literature. From the extensive simulation studies we observe that the proposed test outperforms the existing one with a wide margin in power for moderate sample size. The developed methodology performs satisfactorily for the average daily temperature of central England and monthly global average anomaly of temperatures.
△ Less
Submitted 17 March, 2015;
originally announced March 2015.
-
Bayesian Nonparametric Dynamic State Space Modeling with Circular Latent States
Authors:
Satyaki Mazumder,
Sourabh Bhattacharya
Abstract:
State space models are well-known for their versatility in modeling dynamic systems that arise in various scientific disciplines. Although parametric state space models are well studied, nonparametric approaches are much less explored in comparison. In this article we propose a novel Bayesian nonparametric approach to state space modeling assuming that both the observational and evolutionary funct…
▽ More
State space models are well-known for their versatility in modeling dynamic systems that arise in various scientific disciplines. Although parametric state space models are well studied, nonparametric approaches are much less explored in comparison. In this article we propose a novel Bayesian nonparametric approach to state space modeling assuming that both the observational and evolutionary functions are unknown and are varying with time; crucially, we assume that the unknown evolutionary equation describes dynamic evolution of some latent circular random variable.
Based on appropriate kernel convolution of the standard Wiener process we model the time-varying observational and evolutionary functions as suitable Gaussian processes that take both linear and circular variables as arguments. Additionally, for the time-varying evolutionary function, we wrap the Gaussian process thus constructed around the unit circle to form an appropriate circular Gaussian process. We show that our process thus created satisfies desirable properties.
For the purpose of inference we develop an MCMC based methodology combining Gibbs sampling and Metropolis-Hastings algorithms. Applications to a simulated data set, a real wind speed data set and a real ozone data set demonstrated quite encouraging performances of our model and methodologies.
△ Less
Submitted 22 July, 2015; v1 submitted 13 August, 2014;
originally announced August 2014.
-
Single step and multiple step forecasting in one dimensional single chirp signal using MCMC based Bayesian analysis
Authors:
Satyaki Mazumder
Abstract:
Chirp signals are frequently used in different areas of science and engineering. MCMC based Bayesian inference is done here for purpose of one step and multiple step prediction in case of one dimensional single chirp signal with i.\ i.\ d.\ error structure as well as dependent error structure with exponentially decaying covariances. We use Gibbs sampling technique and random walk MCMC to update th…
▽ More
Chirp signals are frequently used in different areas of science and engineering. MCMC based Bayesian inference is done here for purpose of one step and multiple step prediction in case of one dimensional single chirp signal with i.\ i.\ d.\ error structure as well as dependent error structure with exponentially decaying covariances. We use Gibbs sampling technique and random walk MCMC to update the parameters. We perform total five simulation studies for illustration purpose. We also do some real data analysis to show how the method is working in practice.
△ Less
Submitted 2 July, 2014; v1 submitted 27 June, 2014;
originally announced June 2014.
