"A Hybrid Prediction Model with Time-Varying Gain Tracking Differentiator in Taylor Expansion: Evidence from Precious Metals" by Zhidan Luo

Article

A Hybrid Prediction Model with Time-Varying Gain Tracking Differentiator in Taylor Expansion: Evidence from Precious Metals

Journal of Forecasting (2022)

Zhidan Luo
Wei Guo
Qingfu Liu, Fudan University
Yiuman Tse, University of Missouri-St. Louis

Link

Abstract

In this paper, we propose a modified hybrid prediction model to capture both linear and nonlinear patterns in time-series data by incorporating autoregressive integrated moving average (ARIMA) models and Taylor expansions. We introduce a time-varying gain in the tracking differentiator to reduce the peaking value that occurs in a constant-high-gain design. The models are tested with gold and silver futures prices. The results show that the hybrid model with time-varying high gain tracking differentiator outperforms other hybrid models.

Disciplines

Finance

Publication Date

2022

DOI

10.1002/for.2935

Citation Information

Zhidan Luo, Wei Guo, Qingfu Liu and Yiuman Tse. "A Hybrid Prediction Model with Time-Varying Gain Tracking Differentiator in Taylor Expansion: Evidence from Precious Metals" Journal of Forecasting (2022)
Available at: http://works.bepress.com/yiuman-tse/143/