International Journal of Forecasting最新文献

Estimating financial market volatility is integral to the study of investment decisions and behaviour. Previous literature has, therefore, attempted to identify an optimal volatility forecasting model. However, optimal volatility forecasting is dynamic. It depends on the asset being studied and financial market conditions. We propose a novel empirical methodology to account for this dynamism. Using our Multiple Hypothesis Testing with the False Discovery Rate (FDR) method, we identify buckets of superior-performing models relative to the literature’s benchmark models. We present evidence that our proposed FDR bucket with GJR-GARCH has the lowest forecast error in predicting one-step-ahead realized volatility. We also compare our FDR method with two Family-Wise Error Rate model selection frameworks, and the evidence supports our proposed FDR methodology.

This paper investigates the information content of volatility indices for the purpose of predicting the future direction of the stock market. To this end, different machine learning methods are applied. The dataset used consists of stock index returns and volatility indices of the US stock market from January 2011 until July 2022. The predictive performance of the resulting models is evaluated on the basis of three evaluation metrics: accuracy, the area under the ROC curve, and the F-measure. The results indicate that machine learning models outperform the classical least squares linear regression model in predicting the direction of S&P 500 returns. Among the models examined, random forests and bagging attain the highest predictive performance based on all the evaluation metrics adopted.

This paper introduces general mathematical techniques for stable long-term forecasts with calibrated uncertainty measures. For most time series models, the difficulty of obtaining accurate probabilistic future time step predictions increases with the prediction horizon. We propose a surprisingly simple class of models that characterizes time-varying distributions and enables reasonably accurate predictions thousands of time steps into the future. This technique, called Deep Probabilistic Koopman (DPK), is based on recent advances in linear Koopman operator theory and does not require time stepping for future time predictions. We demonstrate the long-term forecasting performance of these models on a diversity of domains, including electricity demand forecasting, atmospheric chemistry, and neuroscience. Our domain-agnostic technique outperforms all 177 domain-specific competitors in the most recent Global Energy Forecasting Competition for electricity demand modelling.

The paper discusses how standard forecasting tools in multivariate time series analysis are affected when ignoring possible changes in the mean and the (co)variance. We study the estimation, forecasts, and estimated impulse responses of so-called long vector autoregressions, for which the complexity of the model increases with the sample size. We prove that, in spite of structural change in the data generating process, coefficient estimates and out-of-sample forecasts based on such long vector autoregressions are consistent. The sampling behaviour of estimated impulse responses depends primarily on the residual covariance matrix, which converges to an “average” covariance matrix in the case of varying (co)variances. Localised estimators (also obtained by means of a suitable long vector autoregression) may be more suitable in this case. Monte Carlo simulations support our theoretical findings. The empirical relevance of the theory is illustrated in two applications: (i) the international dynamics of inflation, and (ii) uncertainty and economic activity.

The Bayesian statistical paradigm provides a principled and coherent approach to probabilistic forecasting. Uncertainty about all unknowns that characterize any forecasting problem – model, parameters, latent states – is able to be quantified explicitly and factored into the forecast distribution via the process of integration or averaging. Allied with the elegance of the method, Bayesian forecasting is now underpinned by the burgeoning field of Bayesian computation, which enables Bayesian forecasts to be produced for virtually any problem, no matter how large or complex. The current state of play in Bayesian forecasting in economics and finance is the subject of this review. The aim is to provide the reader with an overview of modern approaches to the field, set in some historical context, with sufficient computational detail given to assist the reader with implementation.

An increasing number of household and firm surveys ask for subjective probabilities that the inflation rate falls into various outcome ranges. We provide a new measure of the uncertainty implicit in such probabilities. The measure has several advantages over existing methods: It is robust, trivial to implement, requires no functional form assumptions, and is well-defined for all logically possible probabilities. These advantages are particularly relevant when analyzing microdata from extensive consumer surveys. We illustrate the new measure using data from the Survey of Consumer Expectations.

Using a large quarterly macroeconomic dataset for the period 1960–2017, we document the ability of specific financial ratios from the housing market and firms’ aggregate balance sheets to predict GDP over medium-term horizons in the United States. A cyclically adjusted house price-to-rent ratio and the liabilities-to-income ratio of the non-financial non-corporate business sector provide the best in-sample and out-of-sample predictions of GDP growth over horizons of one to five years, based on a wide variety of rankings. Small forecasting models that include these indicators outperform popular high-dimensional models and forecast combinations. The predictive power of the two ratios appears strong during both recessions and expansions, stable over time, and consistent with well-established macro-finance theory.

The U.S. COVID-19 Forecast Hub aggregates forecasts of the short-term burden of COVID-19 in the United States from many contributing teams. We study methods for building an ensemble that combines forecasts from these teams. These experiments have informed the ensemble methods used by the Hub. To be most useful to policymakers, ensemble forecasts must have stable performance in the presence of two key characteristics of the component forecasts: (1) occasional misalignment with the reported data, and (2) instability in the relative performance of component forecasters over time. Our results indicate that in the presence of these challenges, an untrained and robust approach to ensembling using an equally weighted median of all component forecasts is a good choice to support public health decision-makers. In settings where some contributing forecasters have a stable record of good performance, trained ensembles that give those forecasters higher weight can also be helpful.