A weighted multivariate combination time series forecasting model

Mingli Zhu

doi:10.1117/12.2692517

23 August 2023 A weighted multivariate combination time series forecasting model

Mingli Zhu

Proceedings Volume 12784, Second International Conference on Applied Statistics, Computational Mathematics, and Software Engineering (ASCMSE 2023); 127841D (2023) https://doi.org/10.1117/12.2692517
Event: 2023 2nd International Conference on Applied Statistics, Computational Mathematics and Software Engineering (ASCMSE 2023), 2023, Kaifeng, China

Abstract

In order to further improve the accuracy of time series forecasting, this article proposes a combined prediction model based on four traditional prediction models: GM(1,N), BP neural network, partial least squares regression (PLSR), and triple exponential smoothing (TRIX). The combined prediction model is constructed using quadratic nonlinear programming, while the dynamic weighted combined prediction model is constructed using the reciprocal of variance method. Empirical results show that the combined prediction models effectively mitigate the limitations of the four individual models and exhibit good predictive performance. Among them, the dynamic weighted combined prediction model, compared to the fixed weighted combined prediction model, is more suitable for time series data as the weights have time-varying characteristics.

Citation Download Citation

Mingli Zhu "A weighted multivariate combination time series forecasting model", Proc. SPIE 12784, Second International Conference on Applied Statistics, Computational Mathematics, and Software Engineering (ASCMSE 2023), 127841D (23 August 2023); https://doi.org/10.1117/12.2692517

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available