Journal of Forecasting最新文献

This paper puts forward a new and simple method to combine forecasts, which is particularly useful when the forecasts are strongly correlated. It is based on the Mincer Zarnowitz regression, and a subsequent determination using Shapley values of the weights of the forecasts in a new combination. For a stylized case, it is proved that such a Shapley-value-based combination improves upon an equal-weight combination. Simulation experiments and a detailed illustration show the merits of the Shapley-value-based forecast combination.

This paper proposes a method that uses mean square error (MSE) and model confidence set (MCS) as the loss function of back-propagation neural network (BPNN), aiming to train and find a generalized autoregressive conditional heteroskedastic (GARCH) model that has the best forecasting performance of a time series. Combining MSE and the p-value of MCS can not only estimate better parameters for the GARCH models but also find the best GARCH model to forecast the volatility of a time series. Meanwhile, we divide a time series into several parts and use each part as the input of the BPNN. Through the BPNN, each part of the time series will be turned into several forecasting values. These values will be used to calculate the MSE and the p-value of MCS, which will then be used to update the parameters of the BPNN. In the end, we use MCS to choose the best GARCH model among the trained GARCH models and compare this method with maximum likelihood estimation (MLE) and the generalized least squares estimation (GLS). The result shows that the p-value of MCS of the best model estimated by this method is higher than the p-value of MCS of the best model estimated by MLE and GLS. According to the theory of MCS, a model that has a larger p-value does have a better forecasting performance. The method proposed by this paper can provide a potential application of neural network in GARCH model forecasting and estimation.

Agriculture is facing significant challenges in the development of crop yield forecasts, which are important aspects of decision-making at the international, regional, and local levels. The area of agriculture is attracting growing attention because of increasing the demand for food supplies. To ensure future food supplies, crop yield prediction (CYP) provides the best decision-making to assist farmers in agricultural yield forecasting efficiently. Nevertheless, CYP is a difficult endeavor because of the intricacy of the underlying mechanisms and the effect of numerous factors, including weather patterns, soil characteristics, and crop management techniques. In today's era, ensemble learning (EL) approaches have recently demonstrated significant promise for enhancing the reliability and accuracy of CYP. The success of the EL techniques depends on several facts, including how the base learner models are trained and how these are combined. This study provides important insights into the EL techniques for CYP. This paper proposes an expert system model named precise ensemble expert system for crop yield prediction (PEESCYP) to predict the best crop for agricultural land. The proposed PEESCYP model employs multiple imputation by chained equation (MICE) data imputation technique to treat the missing values of the collected dataset, the isolation forest (IF) technique for outlier detection, the ant colony optimization (ACO) technique to perform feature selection, robust scaling (RS) technique to perform data normalization, and the extra tree (ET) is used for classification to overcome the variance and overfitting problem of the single classifiers. The measurements of the proposed PEESCYP model have been collected by means of accuracy, precision, recall, and F-1 score using a prepared dataset, which is collected from International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), and the proposed model is compared with different single-classifier based ML models, EL models, and various existing models available in the literature. The results of this experiment underline that the proposed PEESCYP model outperforms the others.

Forecasting financial distress of corporations is a difficult task in economies undergoing transition, as data are scarce and are highly imbalanced. This research tackles these difficulties by gathering reliable financial distress data in the context of a transition economy and employing the synthetic minority oversampling technique (SMOTE). The study employs seven different models, including linear discriminant analysis (LDA), logistic regression (LR), support vector machines (SVMs), neural networks (NNs), decision trees (DTs), random forests (RFs), and the Merton model, to predict financial distress among publicly traded companies in Vietnam between 2011 and 2021. The first six models use accounting-based variables, while the Merton model utilizes market-based variables. The findings indicate that while all models perform fairly well in predicting results for nondelisted firms, they perform somewhat poorly in predicting results for delisted firms in terms of various measures including balanced accuracy, Matthews correlation coefficient (MCC), precision, recall, and $�{�F�}_{�1�}$ score. The study shows that the models that incorporate both Altman's and Ohlson's variables consistently outperform those that only use Altman's or Ohlson's variables in terms of balanced accuracy. Additionally, the study finds that NNs are generally the most effective models in terms of both balanced accuracy and MCC. The most important variable in Altman's variables as well as the combination of Altman's and Ohlson's variables is “reat” (retained earnings on total assets), whereas “ltat” (total liabilities on total assets) and “wcapat” (working capital on total assets) are the most important variables in Ohlson's variables. The study also reveals that in most cases, the models perform better in predicting results for big firms than for small firms and typically better than in good years than for bad years.

This paper investigates the high-frequency volatility modeling and prediction for crude oil futures in China, a new asset class emerging in recent years. Two volatility measures, the realized variance ($�\mathrm{RV}$) and realized bi-power variations ($�\mathrm{RBV}$) are constructed at various frequencies by virtue of 1-minute crude oil futures prices. The distinctive components of these volatility estimators are further identified to exploit the information contents in the in-sample explanatory power of the realized variance dynamics and the out-of-sample prediction of realized variance across different horizons, leading to four new HAR-RV-type models. First, the empirical results show that the continuous component of the weekly realized variance, representing investors' trading behavior in the medium-term, is the dominant factor driving up volatility trends in China's crude oil futures market over a range of market conditions. Second, the monthly jump component in realized variance presents the significant in-sample explanatory power, and yet marginally improves prediction performance in realized variance during the two out-of-sample periods. Finally, these results are robust toward various market/model setups, over day- and night-trading hours, and across a range of prediction horizons and relative to prediction benchmarks.

Based on a large simulation study, this paper investigates which strategy to adopt in order to choose the most accurate forecasting model for mixed causal-noncausal autoregressions (MAR) data generating processes: always differencing (D), never differencing (L), or unit root pretesting (P). Relying on recent econometric developments regarding forecasting and unit root testing in the MAR framework, the main results suggest that from a practitioner's point of view, the P strategy at the 10% level is a good compromise. In fact, it never departs too much from the best model in terms of forecast accuracy, unlike the L (respectively, D) strategy when the DGP becomes very persistent (respectively, less persistent). This approach is illustrated using recent monthly Brent crude oil price data.