nep-for New Economics Papers
on Forecasting
Issue of 2021‒06‒21
seven papers chosen by
Rob J Hyndman
Monash University

  1. Quantifying time-varying forecast uncertainty and risk for the real price of oil By Knut Are Aastveit; Jamie Cross; Herman K. van Dijk
  2. Forecasting Sovereign Bond Realized Volatility Using Time-Varying Coefficients Model By Barbora Malinska
  3. Commodity Prices and Forecastability of South African Stock Returns Over a Century: Sentiments versus Fundamentals By Afees A. Salisu; Rangan Gupta
  4. Modeling and forecasting production indices using artificial neural networks, taking into account intersectoral relationships and comparing the predictive qualities of various architectures By Kaukin Andrey; Kosarev Vladimir
  5. Financial Vulnerability and Risks to Growth in Emerging Markets By Acharya, Viral V.; Bhadury, Soumya; Surti, Jay
  6. A Bayesian realized threshold measurement GARCH framework for financial tail risk forecasting By Chao Wang; Richard Gerlach
  7. Conditional macroeconomic forecasts: Disagreement, revisions and forecast errors By Glas, Alexander; Heinisch, Katja

  1. By: Knut Are Aastveit (BI Norwegian Business School); Jamie Cross (BI Norwegian Business School); Herman K. van Dijk (Erasmus University Rotterdam)
    Abstract: We propose a novel and numerically efficient quantification approach to forecast uncertainty of the real price of oil using a combination of probabilistic individual model forecasts. Our combination method extends earlier approaches that have been applied to oil price forecasting, by allowing for sequentially updating of time-varying combination weights, estimation of time-varying forecast biases and facets of miscalibration of individual forecast densities and time-varying inter-dependencies among models. To illustrate the usefulness of the method, we present an extensive set of empirical results about time-varying forecast uncertainty and risk for the real price of oil over the period 1974-2018. We show that the combination approach systematically outperforms commonly used benchmark models and combination approaches, both in terms of point and density forecasts. The dynamic patterns of the estimated individual model weights are highly time-varying, reflecting a large time variation in the relative performance of the various individual models. The combination approach has built-in diagnostic information measures about forecast inaccuracy and/or model set incompleteness, which provide clear signals of model incompleteness during three crisis periods. To highlight that our approach also can be useful for policy analysis, we present a basic analysis of profit-loss and hedging against price risk.
    Keywords: Oil price, Forecast density combination, Bayesian forecasting, Instabilities, Model uncertainty
    Date: 2021–06–13
  2. By: Barbora Malinska (Institute of Economic Studies, Faculty of Social Sciences, Charles University, Opletalova 26, 110 00, Prague, Czech Republic)
    Abstract: This paper studies predictability of realized volatility of U.S. Treasury futures using high-frequency data for 2-year, 5-year, 10-year and 30-year tenors from 2006 to 2017. We extend heterogeneous autoregressive model by Corsi (2009) by higher-order realized moments and allow all model coefficients to be time-varying in order to explore dynamics in forecasting power of individual predictors across the term structure. We find realized kurtosis to be valuable predictor across the term structure with robust contribution also in out-of-sample analysis for the shorter tenors. Time-varying coefficient models are found to bring significant out-of-sample forecasting accuracy gain at the short end of the term structure. Further, we detect significant asymmetry in forecasting errors present for all the tenors as the constant-coefficient models were found to generate systemic under-predictions of future realized volatility.
    Keywords: Realized moments, Sovereign bonds, Volatility forecasting, High-frequency data, Time-varying coefficients
    JEL: C32 C53 G17
    Date: 2021–06
  3. By: Afees A. Salisu (Centre for Econometric and Allied Research, University of Ibadan, Ibadan, Nigeria); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield, 0028, South Africa)
    Abstract: We forecast real stock returns of South Africa over the monthly period of 1915:01 to 2021:03 using real oil, gold and silver prices, based on an autoregressive type distributed lag model that controls for persistence and endogeneity bias. Oil price proxies for fundamentals, while gold and silver prices capture sentiments. We find that the metrics for fundamentals and sentiments both predict real stock returns of South Africa, with nonlinearity, modelled by decomposition of these prices into their respective positive and negative counterparts, playing an important role in terms of forecasting when a longer out-of-sample period spanning over three-quarters of a century is used. When compared to fundamentals, sentiments, particularly real gold prices, have a relatively more stronger role to play in forecasting real stock returns. Further, the predictability of stock returns emanating from fundamentals and sentiments is in line with the findings over the same period derived for two other advanced markets namely, the United Kingdom (UK) and the United States (US), but the stock market of another emerging economy, i.e., India covering 1920:08 to 2021:03, unlike South Africa, is found to be completely unpredictable. In other words, South Africa, in terms of its predictability, behaves like a developed stock market. Finally, given the importance of platinum and palladium for South Africa, our forecasting exercise based on their real prices over 1968:01 to 2021:03, depicts strong predictive content for real stock returns, thus again highlighting the importance of behavioral variables. However, these prices do not necessarily contain additional information over what is already available in gold, silver and oil real prices. Our results have important implications for academicians, investors and policymakers.
    Keywords: Commodity prices, real stock returns, emerging and developed markets, forecasting
    JEL: C22 C53 G15 G17 Q02
    Date: 2021–06
  4. By: Kaukin Andrey (Russian Presidential Academy of National Economy and Public Administration); Kosarev Vladimir (Russian Presidential Academy of National Economy and Public Administration)
    Abstract: This paper analyzes the possibilities of using convolutional and recurrent neural networks to predict the indices of industrial production of the Russian economy. Since the indices are asymmetric in periods of growth and decline, it was hypothesized that nonlinear methods will improve the quality of the forecast relative to linear ones.
    Keywords: convolutional neural networks, recurrent neural networks
    Date: 2021–01
  5. By: Acharya, Viral V.; Bhadury, Soumya; Surti, Jay
    Abstract: This paper introduces a new financial vulnerability index for emerging market economies by exploiting key differences in their business cycles relative to those of advanced economies. Information on the domestic price of risk, cost of dollar hedging and market-based measures of bank vulnerability combine to generate indexes significantly more effective in capturing macro-financial vulnerability and stress compared to those based on information in trade and global factors. Our index significantly augments early warning surveillance capacity, as evidenced by out-of-sample forecasting gains around a majority of turning points in GDP growth relative to distributed lag models that are augmented with information from macro-financial indexes that are custom-built to optimize such forecasts.
    Keywords: business cycles; early warning indicators; financial conditions; price of risk; Vulnerability
    JEL: C53 E32 E44
    Date: 2020–06
  6. By: Chao Wang; Richard Gerlach
    Abstract: In this paper, an innovative threshold measurement equation is proposed to be employed in a Realized-GARCH framework. The proposed framework employs a nonlinear threshold regression specification to consider the leverage effect and model the contemporaneous dependence between the observed realized measures and hidden volatility. A Bayesian Markov Chain Monte Carlo method is adapted and employed for the model estimation and forecasting, with its validity assessed via a simulation study. The usefulness of the proposed measurement equation in a Realized-GARCH model has been evaluated via a comprehensive empirical study, by forecasting the 1% and 2.5% Value-at-Risk and Expected Shortfall on six market indices. The proposed framework is shown to be capable of producing competitive tail risk forecasting results, compared to the original Realized-GARCH. Especially, the proposed model is favoured during the high volatility 2008 Global Financial Crisis period.
    Date: 2021–06
  7. By: Glas, Alexander; Heinisch, Katja
    Abstract: Using data from the European Central Bank's Survey of Professional Forecasters, we analyse the role of ex-ante conditioning variables for macroeconomic forecasts. In particular, we test to which extent the heterogeneity, updating and ex-post performance of predictions for inflation, real GDP growth and the unemployment rate are related to assumptions about future oil prices, exchange rates, interest rates and wage growth. Our findings indicate that inflation forecasts are closely associated with oil price expectations, whereas expected interest rates are used primarily to predict output growth and unemployment. Expectations about exchange rates and wage growth also matter for macroeconomic forecasts, albeit less so than oil prices and interest rates. We show that survey participants can considerably improve forecast accuracy for macroeconomic outcomes by reducing prediction errors for external conditions. Our results contribute to a better understanding of the expectation formation process of experts.
    Keywords: assumptions,disagreement,forecast accuracy,forecast revisions,survey forecasts
    JEL: C53 D84 E02 E32
    Date: 2021

This nep-for issue is ©2021 by Rob J Hyndman. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.