Latent factor models for the Chinese commodity futures markets

The rapid growth of Chinese commodity futures markets over the past several decades has created a fertile ground for exploring underlying market dynamics. In this research, we utilize Instrumented Principal Component Analysis (IPCA) alongside the Conditional Autoencoder (CA) method to construct late...

Full description

Saved in:
Bibliographic Details
Published in:Pacific-Basin finance journal Vol. 93; p. 102890
Main Authors: Liu, Yanchu, Zhou, Heyang, Yang, Haisheng
Format: Journal Article
Language:English
Published: Elsevier B.V 01.10.2025
Subjects:
Commodity futures
Conditional autoencoder
Factor model
Instrumented principal component analysis
Machine learning
Conditional autoencoder
Factor model
Instrumented principal component analysis
Machine learning
Commodity futures
ISSN:0927-538X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The rapid growth of Chinese commodity futures markets over the past several decades has created a fertile ground for exploring underlying market dynamics. In this research, we utilize Instrumented Principal Component Analysis (IPCA) alongside the Conditional Autoencoder (CA) method to construct latent factor models tailored to this market. By uncovering hidden patterns and intrinsic characteristics that drive futures prices, our empirical results demonstrate robust out-of-sample predictive accuracy. •We develop IPCA and CA models for Chinese Commodity Futures Markets.•We expand the set of profitable factors available for Chinese Commodity Futures Markets.•Comprehensive explanations with insights for the uncovered factors are provided.
ISSN:0927-538X
DOI:10.1016/j.pacfin.2025.102890