Journal of Forecasting最新文献

Our expectation of stock price trends is the main factor in making a trading strategy or determining the right time to buy or sell a stock. Predictions on stock market prices are a great challenge due to its complexity and the dependence of prices on various economic and non-economic factors. Today, technological advances have led to proposing new methods in this field. Deep learning is one of these methods that has received much attention in recent years. In this study, using deep learning methods and, more specifically, long-short term memory (LSTM) and gated recurrent unit (GRU) networks, we predict the S&P 500 index in three different time frames: daily, weekly, and monthly. We also compare the efficiency of these two methods and examine the effect of choosing different time frames for the input data of these networks. The results indicate that the GRU network outperforms the LSTM network. Also, selecting an appropriate time frame for the input data may improve the network accuracy.

In response to the research demand for forecasting European Union Allowance (EUA) prices, this paper proposes a probabilistic forecasting framework based on a spatiotemporal convolutional neural network. This framework innovatively integrates multidimensional external uncertainty indicators, captures the long-term dependencies of carbon prices through a spatiotemporal convolutional structure, and combines quantile regression with conformal prediction to effectively estimate prediction intervals. Empirical studies demonstrate that the proposed TCN-iTransformer model outperforms existing methods in both point prediction and interval prediction, exhibiting excellent prediction interval coverage probability and normalized average width at different confidence intervals. The Diebold–Mariano (DM) test and ordinary least squares (OLS) regression analysis further validate the predictive advantages of the proposed model. Furthermore, SHAP analysis reveals that the U.S. Treasury yield spread has the most significant impact on EUA price forecasting, while geopolitical risks predominantly exert negative effects. The research findings provide important references for constructing risk mitigation strategies in the European Union carbon emissions market under complex market environments.

We examine the time-varying jump behavior of the Chinese stock market across various time scales. A novel hybrid model (VMD-ARJI-X) is proposed that integrates variational mode decomposition (VMD) with the autoregressive jump intensity model (ARJI), which also incorporates monetary policy and investor sentiment as explicit variables. Results indicate that the interactions between monetary policy and investor sentiment significantly amplify jump risk at the short- and medium-term scales, while the effect is less pronounced over long-term horizons. The predictive capability of the VMD-ARJI-X, which embeds interest rate and investor sentiment into the jump intensity component, outperforms other benchmark models. Our findings provide policymakers with actionable insights for identifying extreme risks triggered by both policy and sentiment and obtaining forward-looking warnings. The multiscale jump signals offer a practical tool to design dynamic risk management strategies for investors.

Accurate prediction of carbon prices is crucial for policymakers, investors, and other participants in emissions trading schemes (ETS) and during regulatory transitions. In this work, carbon price movements are forecasted using a nonlinear ARIMA model as the baseline, alongside XGBoost and LSTM as competing models. The widely adopted XGBoost model is a machine learning (ML) technique, while the LSTM model belongs to the class of Recurrent Neural Network (RNN) models. To harness the predictive strengths of both approaches, we also employ a hybrid model that averages forecasts from the LSTM and XGBoost models. The dataset used in this study is in daily format, ranging from December 1, 2010, to January 10, 2025. The results show that both XGBoost and LSTM outperform the baseline ARIMA model. Furthermore, the hybrid model demonstrates statistically significant improvements in forecasting accuracy compared to the baseline model. These findings suggest that ML- and RNN-based approaches can serve as effective alternatives to traditional statistical and econometric models in carbon pricing forecasting.

The area under the receiver operating characteristic (AUROC) curve is a widely used tool for assessing and ranking global classifier performance. However, because AUROC ignores the scale of predicted probabilities, it can sometimes provide a misleading performance evaluation. To address this limitation, we build on the area under the Kuipers score curve (AUKSC), and reinterpret this metric by extending the traditional ROC curve into a three-dimensional framework that incorporates thresholds, leading to the area of the generalized ROC (AGROC) curve, thus providing a unified measure of classification performance. Through extensive Monte Carlo simulations, we demonstrate that AGROC effectively addresses the limitations of traditional AUROC metrics, offering a more robust tool for ranking probabilistic classifiers by balancing accuracy and probabilistic differentiation. In an empirical application, we show that AGROC accurately identifies recession probabilities derived from various Markov-switching models applied to US GDP growth data, aligning closely with NBER-defined business cycle phases.

Transformer-based models have witnessed remarkable advancements in the domain of time series forecasting. However, significant challenges persist in effectively handling large volumes of historical data and comprehensively capturing multiscale characteristics inherent in time series. This paper proposes a novel time series forecasting model that integrates the Discrete Wavelet Transform (DWT) and residual learning modules. This integration is aimed at enhancing the model's proficiency in capturing the intricate nonlinear and multiscale features of time series data. The proposed model leverages DWT to decompose the time series into multiple scales, enabling it to effectively capture both local and global features across diverse temporal resolutions. The residual learning modules are meticulously designed to improve the training stability of the model and augment its feature extraction capabilities. Additionally, local and global attention mechanisms are employed to comprehensively capture short- and long-term dependencies within time series data. Comprehensive experiments conducted on seven real-world datasets demonstrate that the proposed approach outperforms state-of-the-art deep learning models in long-term time series forecasting tasks. It achieves higher accuracy and better generalization performance. Ablation studies are also carried out, which further validate the individual contributions of each module to the overall performance of the proposed model, providing strong evidence for the effectiveness of the model's design.

Tracking house prices is crucial for identifying risks to the banking sector and overall financial stability, making accurate predictions essential. This study examines whether housing-related stock returns can predict house price fluctuations in the United Kingdom. Using monthly data from 1983 to 2023, empirical evidence suggests that these equity returns strongly predict UK house price changes 1 month ahead. Because housing-related stock prices provide reliable and easily accessible forecasts of housing market trends, the findings offer valuable insights for investors and policymakers.

In the area of econometrics, the investigation and characterization of processes that retain memory for the past are often of interest. This work overcomes collinearity problems that arise in distributed lag formulations by modeling these effects as structural elements within nonlinear dynamic models using transfer functions. Our main contribution lies in performing sequential Bayesian inference for nonlinear dynamic models, providing an efficient computational solution based on analytical approximations. The scalability offered by the proposed sequential method is particularly relevant in the econometric context, where long time series or multiple levels of disaggregation are often encountered. The proposed models incorporate stochastic volatility, achieved through the use of discount factors. An extensive simulation investigation validates the inferential approximation. The results of the proposed sequential and analytical approximation are compared with the inference obtained through Hamiltonian Monte Carlo in a particular application to real-world consumption data. The results show that the sequential approach produces results that are largely comparable while requiring a significantly shorter amount of computing time. Using the proposed Bayesian state-space framework and a thorough examination of the Phillips curve, a case study is developed focusing on the relationship between inflation and the output gap in the Brazilian scenario. We conclude with a substantial contribution, based on an innovative approach that preserves Bayesian sequential inference and offers a joint model for inflation and the output gap, with dynamic predictive structures assigned to the means, precisions, and correlation between both economic indicators.