Statistical Deep Learning For Assets Forecast

Abstract

Recently, global attention has been paid to climate change. On this account, the market-based carbon pricing scheme is developed to limit greenhouse gas emissions, where a proper grasp of the pricing mechanism is crucial for alleviating global warming. Accordingly, we propose a novel method to interpret carbon price dynamics, concurrently deriving the precise prediction and causality. Due to the nonlinearity and nonstationarity of carbon prices, we develop a real-time decomposition approach coupling the multiple ensemble patch transform (MEPT) and the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN). The MEPT captures the multi-resolution trends of the carbon prices series exactly, and then the ICEEMDAN extracts the fluctuation patterns. Additionally, we collect the numerous potential factors, involving energy sources, energy prices, stock market indices, and economic information. Furthermore, we developed causal temporal convolutional networks (CTCNs) to realize the accurate prediction and the proper causal inference simultaneously. The experimental results on the European Union Allowance (EUA) confirm the effectiveness and necessity of the real-time MEPT-ICEEMDAN decomposition. Moreover, the proposed MEPT-ICEEMDAN-CTCN model exhibits significant superiority in multi-step-ahead and quantile forecast, which realizes the 0.73881%, 1.04461%, and 1.23495% MAPE in one-, five-, and ten-step-ahead forecast respectively and 0.00032 PD_Q^0.1 and the 0.00285 PD_Q^0.9 in the quantile forecast. Meanwhile, it reveals the nonlinear Granger causality across the various horizons and quantiles for the first time.

Furthermore, we construct the unsupervised temporal convolutional networks (UTCNs) for the unsupervised learning of Expected Shortfall (ES) and Value-at-Risk (VaR), and causal UTCNs (CUTCNs) for the causal unsupervised learning, simultaneously driving the Granger causality for the tail risks. Our proposed UTCN models exactly forecast the tail risk of daily returns on four international equity indices, cutting the average losses on test sets from above 0.74380 to lower than -3.27880. Meanwhile, the CUTCN model yields the accurate forecast and nonlinear causality of carbon futures tail risk concurrently, reducing the average losses from 1.96149 to -2.38459 and revealing the causality from stock indices, energy source, and macroeconomic situation to tail risks of EUA.

It is instructive and inspiring for policymakers, industries, investors, and researchers.

Date of Award	3 Aug 2023
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Min Chen (External Supervisor) & Qi WU (Supervisor)