Authors: Sudama Iyer, Shambhu Chandra, Vivek Tripathi, Jagnarayana Thakur, Ajaharu Din
Abstract: Time series analysis and stochastic process theory form the backbone of modern statistical modeling for data observed over time and under uncertainty. These methods are widely applied in engineering, economics, finance, environmental science, signal processing, and biological systems. The main objective of this paper is to present a comprehensive review of time series and stochastic process methods, highlighting their theoretical foundations, commonly used models, estimation techniques, and recent methodological developments. Classical approaches such as autoregressive moving average (ARMA) models, spectral analysis, and Markov processes are discussed alongside advanced techniques including state space models, stochastic differential equations, and non-linear time series models. The paper also emphasizes practical applications and challenges in real-world data analysis, such as non-stationarity, noise, and missing observations. Although the paper is primarily a review, it aims to provide intuitive explanations and comparative insights that may be useful for researchers and practitioners working in applied domains.
Keywords: Time series analysis, stochastic processes, ARIMA models, Markov processes, state space models, random processes, forecasting
Full Issue
| View or download the full issue | PDF 53-63 |