International Journal of Forecasting最新文献

This study develops a shrinkage method, LASSO with a Markov regime-switching model (MRS-LASSO), to predict US stock market volatility. A set of 17 well-known macroeconomic and financial factors are used. The out-of-sample results reveal that the MRS-LASSO model yields statistically and economically significant volatility predictions. We further investigate the predictability of MRS-LASSO with respect to different market conditions, business cycles, and variable selection. Three factors (equity market returns, a short-term reversal factor, and a consumer sentiment index) are the most frequent predictors. To investigate the practical implications, we construct the expected variance risk premium (VRP) by using volatility forecasts generated from the LASSO and MRS-LASSO models to forecast future stock returns and find that those models are also powerful.

Standard methods for forecasting electricity loads are not robust to cyberattacks on electricity demand data, potentially leading to severe consequences such as major economic loss or a system blackout. Methods are required that can handle forecasting under these conditions and detect outliers that would otherwise go unnoticed. The key challenge is to remove as many outliers as possible while maintaining enough clean data to use in the regression. In this paper we investigate robust approaches with data-driven tuning parameters, and in particular present an adaptive trimmed regression method that can better detect outliers and provide improved forecasts. In general, data-driven approaches perform much better than their fixed tuning parameter counterparts. Recommendations for future work are provided.

We examine the internal consistency of US households’ inflation expectations reported as point and density forecasts by the New York Fed’s Survey of Consumer Expectations. We find that the majority of the households report well-defined histograms, with their central tendencies close to the corresponding point forecasts. We observe higher levels of consistency in forecasts reported by survey respondents with higher levels of income, education, and financial literacy. Furthermore, our results suggest that both the point forecasts directly reported and those derived from the histograms are more accurate when they are from respondents who are more likely to report consistent forecasts. In addition, we find that the consensus derived using only the consistent forecasts is as accurate as the consensus derived using all forecasts.

Credit scoring is an important tool to guard against commercial risks for banks and lending companies and provides good conditions for the construction of individual personal credit. Ensemble algorithms have shown appealing progress for the improvement of credit scoring. In this study, to meet the challenge of large-scale credit scoring, we propose a heterogeneous deep forest model (Heter-DF), which is established based on considerations ranging from base learner selection, encouragement of the diversity of base learners, and ensemble strategies, for credit scoring. Heter-DF is designed as a scalable cascading framework that can increase its complexity with the scale of the credit dataset. Moreover, each level of Heter-DF is built by multiple heterogeneous tree-based ensembled base learners, avoiding the homogeneous prediction of the ensemble framework. In addition, a weighted voting mechanism is introduced to highlight important information and suppress irrelevant features, making Heter-DF a robust model for credit scoring. Experimental results on four credit scoring datasets and six evaluation metrics show that the cascading framework a good choice for the ensemble of tree-based base learners. A comparison among homogeneous ensembles and heterogeneous ensembles further demonstrates the effectiveness of Heter-DF. Experiments on different training sets indicate that Heter-DF is a scalable framework which not only deals with large-scale credit scoring but also satisfies the condition where small-scale credit scoring is desirable. Finally, based on the good interpretability of a tree-based structure, the global interpretation of Heter-DF is preliminarily explored.

External sector surveillance and stabilization are core missions of the International Monetary Fund (IMF). Since 1992, the IMF approved over 600 crisis country loan programs, conditional on reforms and performance targets that are contingent on IMF crisis assessments and recovery forecasts. The literature evaluating IMF crisis forecasts has primarily focused on GDP, inflation, and fiscal budgets, but IMF programs often originate with the balance of payments crises. Our evaluation of IMF imports/exports/exchange rates in crisis countries reveals a surprising dichotomy: import forecasts are largely efficient and unbiased, while exports and exchange rate forecasts exhibit substantial biases and inefficiencies. We show forecast errors in the full sample are driven by deeply flawed IMF forecasts for LICs in crisis. Fixed exchange rate LICs (predominantly African franc zone countries) receive systematically inefficient import forecasts. Exchange rate forecasts for LICs with flexible exchange rates are so inefficient that they cannot outperform a naïve random walk, and over 30 percent of the forecasts cannot match the exchange rate’s directional movement during the first year of the recovery. Examining the sources of biases and inefficiencies, we highlight effects of conditionality and geopolitics that were not fully accounted for in IMF forecasts, specifically those relating to arrears (domestic and foreign), fiscal finance (balance and credit limits), policy reforms (trade and government), (civil) wars, and elections.

Forecasting the analyst rating quality (ARQ), defined as whether a stock rating provided by an analyst can correctly foretell the stock movement, is crucial to fully leveraging the value of this information resource. This study develops a two-phase method to identify key predictors for ARQ forecasting. In the first stage, we conduct a thorough literature review to obtain a comprehensive list of candidate features, and organise them under three categories: analyst-related, rating-related, and stock-related. In the second stage, we propose a heterogeneous community-based ensemble feature selection method (ComEFS), with the goal of identifying a subset of relevant predictors to be jointly used for ARQ forecasting. Thorough experiments are conducted on a real dataset to verify the effectiveness of our proposed method. The empirical results show that key predictors identified by ComEFS exhibit stronger predictive power compared to those identified by benchmark methods. This study provides insights about ARQ forecasting by selecting the right input. Selectively utilizing these predictive features can help improve the performance of downstream machine learning models and ultimately help investors avoid unreliable analyst ratings and financial loss.

Dynamic linear models with discounting are state-space models that are sufficiently flexible, interpretable, and computationally efficient. As such they are increasingly applied in economics and finance. Successful modelling and forecasting with such models depends on an appropriate choice of the discount factor. In this work we develop an adaptive approach to sequentially estimate this parameter, which relies on the minimisation of the one-step-ahead forecast error. Simulated data and an in-depth empirical application to the problem of forecasting quarterly UK house prices show that our approach can significantly improve forecast accuracy at a computational cost that is orders of magnitude smaller than approaches based on sequential Monte Carlo. We also conduct an extensive evaluation of diverse forecast combination methods for the task of predicting UK house prices. Our results indicate that a recent density combination method can substantially improve forecast accuracy.

The theoretical advances in the properties of scoring rules over the past decades have broadened the use of scoring rules in probabilistic forecasting. In meteorological forecasting, statistical postprocessing techniques are essential to improve the forecasts made by deterministic physical models. Numerous state-of-the-art statistical postprocessing techniques are based on distributional regression evaluated with the continuous ranked probability score (CRPS). However, the theoretical properties of such evaluations with the CRPS have solely considered the unconditional framework (i.e. without covariates) and infinite sample sizes. We extend these results and study the rate of convergence in terms of the CRPS of distributional regression methods. We find the optimal minimax rate of convergence for a given class of distributions and show that the $k$-nearest neighbor method and the kernel method reach this optimal minimax rate.