Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Econometrics›Structural Time Series Model (Basic Structural Model)
Regression model

Structural Time Series Model (Basic Structural Model)

Basic Structural Model (Structural Time Series Model) · Also known as: BSM, basic structural model, unobserved components model, Yapısal Zaman Serisi Modeli (BSM)

The Structural Time Series Model, in its Basic Structural Model (BSM) form, is Andrew Harvey's state-space approach that decomposes a series into separate stochastic trend, seasonal, cyclical, and irregular components. Developed in Harvey's 1990 treatment, it is prized for interpretability and component decomposition where ARIMA only delivers a black-box fit.

ScholarGate
  1. Regression model
  2. v1
  3. 2 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Structural Time Series Model
ARIMABayesian Structural Time…Markov-Switching ModelVAR ModelETS ModelExponential SmoothingHolt-WintersProphetState Space Model

When to use it

Use the BSM for a continuous time series (at least about 30 observations) when you want an interpretable decomposition into trend, seasonal, and cyclical parts rather than just a forecast. It assumes the series can be written in state-space form, that the Kalman filter is applicable, and that the components are meaningful and interpretable. It is a strong choice when you need to read off the underlying trend or seasonal pattern, handle missing observations, or explain a forecast to non-specialists; it is less necessary when a plain ARIMA fit suffices and interpretability is not a concern.

Strengths & limitations

Strengths
  • Decomposes the series into separate, directly interpretable trend, seasonal, cyclical, and irregular components.
  • Often more interpretable than ARIMA because each component carries an explicit meaning rather than abstract autoregressive and moving-average terms.
  • The state-space and Kalman-filter formulation handles missing observations and time-varying structure naturally.
Limitations
  • Requires casting the problem in state-space form and choosing which components (trend, seasonal, cycle) to include.
  • Maximum-likelihood estimation of the component variances can be sensitive to starting values and the optimisation routine.
  • Needs a reasonable run of data (at least about 30 observations) and well-defined, interpretable components to be reliable.

Frequently asked

How does the BSM differ from ARIMA?

ARIMA describes a series through abstract autoregressive and moving-average terms after differencing, which fits well but is hard to interpret. The BSM instead models explicit trend, seasonal, cyclical, and irregular components, so you can read each part directly. For decomposition and interpretability the BSM is generally preferred.

What role does the Kalman filter play?

The model is written in state-space form, and the Kalman filter recursively estimates the unobserved components as each observation arrives, updating its prediction by a correction proportional to the prediction error. It also produces the likelihood used to estimate the component variances.

What does the 'local linear trend' mean?

It is a trend with both a level and a slope that are each allowed to drift stochastically over time, so the trend can change its direction gradually rather than being fixed. It is the standard trend specification in the Basic Structural Model.

How much data do I need?

At least about 30 observations, and enough span to identify any seasonal and cyclical components you include; otherwise their variances can become unstable or collapse to zero.

Sources

  1. Harvey, A. C. (1990). Forecasting, Structural Time Series Models and the Kalman Filter. Cambridge University Press. ISBN: 978-0521405737
  2. Harvey, A. C. & Shephard, N. (1993). Structural Time Series Models. In G. S. Maddala, C. R. Rao & H. D. Vinod (Eds.), Handbook of Statistics, Vol. 11 (pp. 261-302). Elsevier. DOI: 10.1016/S0169-7161(05)80045-8 ↗

How to cite this page

ScholarGate. (2026, June 1). Basic Structural Model (Structural Time Series Model). ScholarGate. https://scholargate.app/en/econometrics/structural-time-series

Related methods

ARIMABayesian Structural Time SeriesMarkov-Switching ModelVAR Model

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • ARIMAEconometrics↔ compare
  • Bayesian Structural Time SeriesBayesian↔ compare
  • Markov-Switching ModelEconometrics↔ compare
  • VAR ModelEconometrics↔ compare
Compare side by side →

Referenced by

ETS ModelExponential SmoothingHolt-WintersProphetState Space Model

Similar methods

State Space ModelTime Series Kalman FilterBayesian Structural Time SeriesTime-varying parameter SARIMA modelTime series Bayesian inferenceETS ModelTime-varying parameter ARIMA modelSTL Decomposition

Related reference concepts

Hidden Markov ModelsStructural and Latent Variable ModelsTime-Series Models • Dynamic Quantile Regressions • Dynamic Treatment Effect Models • Diffusion Processes • State Space ModelsMultilevel and Partial Pooling ModelsStructural Equation ModelingBayesian Model Averaging

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Structural Time Series Model (Basic Structural Model (Structural Time Series Model)). Retrieved 2026-07-21 from https://scholargate.app/en/econometrics/structural-time-series · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Andrew C. Harvey
Year
1990
Type
State-space (unobserved components) time series model
Estimator
Kalman filter with maximum likelihood
Components
trend, seasonal, cycle, irregular
Outcome
continuous time series
Related methods
ARIMABayesian Structural Time SeriesMarkov-Switching ModelVAR Model
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account