Time series

Prado presents Bayesian methods for analysing time series, pairing established modelling techniques with recent advances in both methodology and applied work. Written at graduate level, the book surveys state-of-the-art approaches to single-variable and multi-variable series, along with newer research directions.

The chapters outline several families of models and the methods used to estimate them, including statistical computation for fitting and checking models and for producing forecasts. The book also links time-domain and frequency-domain analysis. Readers learn to build and test these models with simulation tools such as Markov chain Monte Carlo and sequential Monte Carlo. Worked examples and case studies draw on signal processing, biomedicine and finance. Datasets and code in R and MATLAB are provided online.

Beyond the core material, the text introduces more advanced techniques for difficult problems and shows how time series analysis is expanding into new fields.