Bayesian Structural Time Series

Almarashi, Abdullah M.; Khan, Khushnoor

doi:10.1166/nnl.2020.3083

Cited by 14 publications

(11 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The working of the state space models is Markovian in nature, as the future stage depends on the present state consequently, and computations are iterative. In this respect, several time series models have been developed, including structural time series, state space models, Kalman filter models, and dynamic linear models [22]. Bayesian Structural Time Series (BSTS) model was proposed by [23,24], which is a technique that can be used for selection of the features, forecasting of time series, deducing any causal relationship [22].…”

Section: Bayesian Structural Time Seriesmentioning

confidence: 99%

Prediction of the Demand Response Behavior of Residential Consumers through a Bayesian Forecasting Method

Kaheh

Arasteh

2023

Preprint

View full text Add to dashboard Cite

Demand-side resources have been recognized as a cost-effective approach for providing the required flexibility in power systems with high penetrations of renewable resources. Almost all existing demand response (DR) models are established based on the assumption that end-users are always rational and predictable agents with well-defined preferences. This assumption has resulted in apparently unaccountable divergence between modelled and observed results for DR approaches. However, due to the nature of the prompt responses of DR providers to DR signals, it is of paramount importance to consider the effect of impulsive decisions on providing the flexibility. This paper investigates the effects of impulsive decisions on the frequency of being unresponsive to DR signals and the collaboration rate in 5-minutes ahead DRX through a Bayesian Forecasting Method. The analyses have demonstrated considering the provided flexibility in similar days of week as explanatory variable in Bayesian structural time series can be helpful for the detection of impulsive behavior to attain more accurate DR behavior forecasting.

show abstract

Section: Bayesian Structural Time Seriesmentioning

confidence: 99%

Prediction of the Demand Response Behavior of Residential Consumers through a Bayesian Forecasting Method

Kaheh

Arasteh

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Various methods of applying Bayesian structural time series make it possible to use them to build models taking into account various prior distributions, reduce uncertainty when choosing a model, and work with different types of distributions [49,50], [51,52], [53,54], [55].…”

Section: Analysis Of Literary Datamentioning

confidence: 99%

“…The structural time series model can be described by a pair of equations [51]. The first, the observation equation, relates the observed data 𝑦 𝑡 to a vector of latent variables 𝛼 𝑡 , which is called the "state".…”

Section: Features Of Building Structural Models Of Time Seriesmentioning

confidence: 99%

“…As an example, we present the basic structural model of a time series with predictors as follows [51]: Bayesian Structural Time Series (BSTS) is related to the linear Gaussian model that is used in Kalman filters. The BSTS model is based on more complex mathematical principles than those used in the linear Gaussian model.…”

Section: Features Of Building Structural Models Of Time Seriesmentioning

confidence: 99%

“…The selected models are used for forecasting. The Software engineering and systems analysis ISSN 2617-4316 (Print) ISSN 2663-7723 (Online) table 1 presents the various components of the BSTS models, using the structure of alternative predictive models [49][50][51][52].…”

Section: Features Of Building Structural Models Of Time Seriesmentioning

confidence: 99%

See 2 more Smart Citations

Modeling and forecasting of nonlinear nonstationary processes based on the Bayesian structural time series

Kalinina¹,

Gozhyj²

2022

AAIT

View full text Add to dashboard Cite

The article describes an approach to modelling and forecasting non-linear non-stationary time series for various purposes using Bayesian structural time series. The concepts of non-linearity and non-stationarity, as well as methods for processing non-linearity’sand non-stationarity in the construction of forecasting models are considered. The features of the Bayesian approach in the processing of nonlinearities and nonstationaryare presented. An approach to the construction of probabilistic-statistical models based on Bayesian structural models of time series has been studied. Parametric and non-parametric methods for forecasting non-linear and non-stationary time series are considered. Parametric methods include methods: classical autoregressive models, neural networks, models of support vector machines, hidden Markov models. Non-parametric methods include methods: state-space models, functional decomposition models, Bayesian non-parametric models. One of the types of non-parametric models isBayesian structural time series. The main features of constructing structural time series are considered. Models of structural time series are presented. The process of learning the Bayesianstructural model of time series is described. Training is performed in four stages: setting the structure of the model and a priori probabilities; applying a Kalman filter to update state estimates based on observed data;application of the “spike-and-slab”method to select variables in a structural model; Bayesian averaging to combine the results to make a prediction. An algorithm for constructing a Bayesian structural time seriesmodel is presented. Various components of the BSTS model are considered andanalysed, with the help of which the structures of alternative predictive models are formed. As an example of the application of Bayesian structural time series, the problem of predicting Amazon stock prices is considered. The base dataset is amzn_share. After loading, the structure and data types were analysed, and missing values were processed. The data are characterized by irregular registration of observations, which leads to a large number of missing values and “masking” possible seasonal fluctuations. This makes the task of forecasting rather difficult. To restore gaps in the amzn_sharetime series, the linear interpolation method was used. Using a set of statistical tests (ADF, KPSS, PP), the series was tested for stationarity. The data set is divided into two parts: training and testing. The fitting of structural models of time series was performed using the Kalman filterand the Monte Carlo method according to the Markov chain scheme. To estimate and simultaneously regularize the regression coefficients, the spike-and-slab method was applied. The quality of predictive models was assessed.

show abstract

The Perspective of the Creative Economy Stimulus on the Thai Economy: Explication by BSTS Mixed with the CGE Model

Chokethaworn,

Chaiboonsri,

Eakkapun

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Bayesian Structural Time Series

Cited by 14 publications

References 0 publications

Prediction of the Demand Response Behavior of Residential Consumers through a Bayesian Forecasting Method

Prediction of the Demand Response Behavior of Residential Consumers through a Bayesian Forecasting Method

Modeling and forecasting of nonlinear nonstationary processes based on the Bayesian structural time series

The Perspective of the Creative Economy Stimulus on the Thai Economy: Explication by BSTS Mixed with the CGE Model

Contact Info

Product

Resources

About