nep-for New Economics Papers
on Forecasting
Issue of 2018‒04‒16
eight papers chosen by
Rob J Hyndman
Monash University

  1. Does the choice of estimator matter for forecasting? A revisit By Afees A. Salisu; Ahamuefula Ephraim Ogbonna; Paul Adeoye Omosebi
  2. DSGE forecasts of the lost recovery By Cai, Michael; Del Negro, Marco; Giannoni, Marc; Gupta, Abhi; Li, Pearl; Moszkowski, Erica
  3. Euro area real-time density forecasting with financial or labor market frictions By McAdam, Peter; Warne, Anders
  4. On the robustness of the principal volatility components By Trucíos Maza, Carlos César; Hotta, Luiz Koodi; Pereira, Pedro L. Valls
  5. Tests for Forecast Instability and Forecast Failure under a Continuous Record Asymptotic Framework By Alessandro Casini
  6. Prediction bands for solar energy: New short-term time series forecasting techniques By Michel Fliess; Cédric Join; Cyril Voyant
  7. Could this be a fiction? Bitcoin forecasts most tradable currency pairs better than ARFIMA By Afees A. Salisu; Lateef O. Akanni; Rasheed O. Azeez
  8. Forecasting the Real Index of Gross Domestic Product Using Dynamic Factor Models By Pleskachev, Yury; Ponomarev, Yury

  1. By: Afees A. Salisu (Centre for Econometric and Allied Research, University of Ibadan); Ahamuefula Ephraim Ogbonna (Centre for Econometric and Allied Research, University of Ibadan Department of Statistics, University of Ibadan, Ibadan, Nigeria); Paul Adeoye Omosebi (Centre for Econometric and Allied Research, University of Ibadan. Department of Computer Sciences, University of Lagos, Akoka, Nigeria.)
    Abstract: In this study, we further examine whether the choice of estimator matters for forecasting based on the conclusion of Westerlund and Narayan [WN, hereafter] (2012, 2015). A similar but small simulation study was conducted by WN (2012, 2015) to validate the need to account for salient features of predictors such as persistence, endogeneity and conditional heteroscedasticity in a forecast model. In addition to considering a more representative number of observations for high frequency, extensive replications and four competing estimators, we offer alternative functions for these effects and thereafter, we test whether the conclusion of WN (2012, 2015) will still hold. Our results further lend support to the WN (2012, 2015) findings and thus suggest that the choice of estimator matters for forecasting notwithstanding the alternative functions and scenarios considered in our study. Thus, pre-testing the predictors in a forecast model for the mentioned features is required to identify the appropriate estimator to apply.
    Keywords: Endogeneity, Heteroscedasticity, Persistence, Forecast evaluation
    JEL: C15 C52 C53
    Date: 2018–04
  2. By: Cai, Michael (Federal Reserve Bank of New York); Del Negro, Marco (Federal Reserve Bank of New York); Giannoni, Marc (Federal Reserve Bank of Dallas); Gupta, Abhi (Federal Reserve Bank of New York); Li, Pearl (Federal Reserve Bank of New York); Moszkowski, Erica (Harvard Business School)
    Abstract: The years following the Great Recession were challenging for forecasters. Unlike other deep downturns, this recession was not followed by a swift recovery, but generated a sizable and persistent output gap that was not accompanied by deflation as a traditional Phillips curve relationship would have predicted. Moreover, the zero lower bound and unconventional monetary policy generated a policy environment without precedents. We document the real real-time forecasting performance of the New York Fed dynamic stochastic general equilibrium (DSGE) model during this period and explain the results using the pseudo real-time forecasting performance results from a battery of DSGE models. We find the New York Fed DSGE model’s forecasting accuracy to be comparable to that of private forecasters—and notably better, for output growth, than the median forecasts from the FOMC’s Summary of Economic Projections. The model’s financial frictions were key in obtaining these results, as they implied a slow recovery following the financial crisis.
    Keywords: DSGE models; real-time forecasts; Great Recession; financial frictions
    JEL: C11 C32 C54 E43 E44
    Date: 2018–03–01
  3. By: McAdam, Peter; Warne, Anders
    Abstract: We compare real-time density forecasts for the euro area using three DSGE models. The benchmark is the Smets-Wouters model and its forecasts of real GDP growth and inflation are compared with those from two extensions. The first adds financial frictions and expands the observables to include a measure of the external finance premium. The second allows for the extensive labor-market margin and adds the unemployment rate to the observables. The main question we address is if these extensions improve the density forecasts of real GDP and inflation and their joint forecasts up to an eight-quarter horizon. We find that adding financial frictions leads to a deterioration in the forecasts, with the exception of longer-term inflation forecasts and the period around the Great Recession. The labor market extension improves the medium to longer-term real GDP growth and shorter to medium-term inflation forecasts weakly compared with the benchmark model. JEL Classification: C11, C32, C52, C53, E37
    Keywords: Bayesian inference, DSGE models, forecast comparison, inflation, output, predictive likelihood
    Date: 2018–04
  4. By: Trucíos Maza, Carlos César; Hotta, Luiz Koodi; Pereira, Pedro L. Valls
    Abstract: In this paper, we analyse the recent principal volatility components analysis procedure. The procedure overcomes several diculties in modelling and forecasting the conditional covariance matrix in large dimensions arising from the curse of dimensionality. We show that outliers have a devastating e↵ect on the construction of the principal volatility components and on the forecast of the conditional covariance matrix and consequently in economic and financial applications based on this forecast. We propose a robust procedure and analyse its finite sample properties by means of Monte Carlo experiments and also illustrate it using empirical data. The robust procedure outperforms the classical method in simulated and empirical data.
    Date: 2018–03
  5. By: Alessandro Casini
    Abstract: We develop a novel continuous-time asymptotic framework for inference on whether the predictive ability of a given forecast model remains stable over time. We formally define forecast instability from the economic forecaster's perspective and highlight that the time duration of the instability bears no relationship with stable period. Our approach is applicable in forecasting environment involving low-frequency as well as high-frequency macroeconomic and financial variables. As the sampling interval between observations shrinks to zero the sequence of forecast losses is approximated by a continuous-time stochastic process (i.e., an Ito semimartingale) possessing certain pathwise properties. We build an hypotheses testing problem based on the local properties of the continuous-time limit counterpart of the sequence of losses. The null distribution follows an extreme value distribution. While controlling the statistical size well, our class of test statistics feature uniform power over the location of the forecast failure in the sample. The test statistics are designed to have power against general form of insatiability and are robust to common forms of non-stationarity such as heteroskedasticty and serial correlation. The gains in power are substantial relative to extant methods, especially when the instability is short-lasting and when occurs toward the tail of the sample.
    Date: 2018–03
  6. By: Michel Fliess (AL.I.E.N. - ALgèbre pour Identification & Estimation Numériques, LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau] - CNRS - Centre National de la Recherche Scientifique - Polytechnique - X); Cédric Join (AL.I.E.N. - ALgèbre pour Identification & Estimation Numériques, CRAN - Centre de Recherche en Automatique de Nancy - UL - Université de Lorraine - CNRS - Centre National de la Recherche Scientifique, NON-A - Non-Asymptotic estimation for online systems - Inria Lille - Nord Europe - Inria - Institut National de Recherche en Informatique et en Automatique - CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189 - Université de Lille, Sciences et Technologies - Ecole Centrale de Lille - Inria - Institut National de Recherche en Informatique et en Automatique - Université de Lille, Sciences Humaines et Sociales - Institut Mines-Télécom [Paris] - CNRS - Centre National de la Recherche Scientifique); Cyril Voyant (SPE - Sciences pour l'environnement - UPP - Université Pascal Paoli - CNRS - Centre National de la Recherche Scientifique, Centre hospitalier d'Ajaccio)
    Abstract: Short-term forecasts and risk management for photovoltaic energy is studied via a new standpoint on time series: a result published by P. Cartier and Y. Perrin in 1995 permits, without any probabilistic and/or statistical assumption, an additive decomposition of a time series into its mean, or trend, and quick fluctuations around it. The forecasts are achieved by applying quite new estimation techniques and some extrapolation procedures where the classic concept of "seasonalities" is fundamental. The quick fluctuations allow to define easily prediction bands around the mean. Several convincing computer simulations via real data, where the Gaussian probability distribution law is not satisfied, are provided and discussed. The concrete implementation of our setting needs neither tedious machine learning nor large historical data, contrarily to many other viewpoints.
    Keywords: mean,quick fluctuations,time series,prediction bands,short-term forecasts,Solar energy,persistence,risk,volatility,normality tests
    Date: 2018
  7. By: Afees A. Salisu (Centre for Econometric and Allied Research, University of Ibadan); Lateef O. Akanni (Department of Economics, University of Lagos,Akoka, Lagos, Nigeria); Rasheed O. Azeez (Department of Economics, University of Ibadan, Nigeria.; Department of Economics, Fountain University, Nigeria.)
    Abstract: In this paper, we attempt to exploit any inherent useful information in Bitcoin to predict the future path of the most tradable currency pairs in the world. We also verify whether the forecast outcomes can compare favourably with the time series model such as the fractionally integrated autoregressive moving average (ARFIMA) model. We follow the Lewellen (2004) and Westerlund and Narayan (2102, 2015) approaches that account for any statistical effect that could bias the regression estimates. Our results suggest that Bitcoin is a good predictor of the selected currency pairs and more importantly, its forecast results outperform the time series model judging by the Diebold and Mariano test regardless of the data sample and forecast horizon. Although, recent evidence in the literature seems to suggest that the Bitcoin bubble will soon burst, its connection with the considered currency pairs may be exploited while it lasts.
    Keywords: Bitcoin, Exchange rates, Forecast evaluation
    JEL: F31 F37 G15
    Date: 2018–03
  8. By: Pleskachev, Yury (Russian Presidential Academy of National Economy and Public Administration (RANEPA)); Ponomarev, Yury (Russian Presidential Academy of National Economy and Public Administration (RANEPA))
    Abstract: Successful implementation of economic policy measures largely depends on the efficiency and accuracy of forecasts of key macroeconomic parameters of the economy. Deceleration of Russia's economic growth in the years 2013-2014 and the subsequent fall in GDP in 2015-2016 once again demonstrated the importance of taking urgent and at the same time balanced decisions, the basis for which should be the most relevant statistical base on key indicators, including real GDP. At the same time quarterly data on GDP dynamics is published with a considerable delay, which leads to the need of short-term forecasts in real time. The use of dynamic factor models for rapid forecasting of GDP has become particularly popular in world literature and also in practice (such models are used by the central Banks of the world's leading countries) over the last few years due to more accurate forecasts that allow to obtain the model data, as well as the fact that they allow to take into account changes in economic conditions and their impact on the country's economy in the formation of PUBLIC policy measures before the relevant statistical data are published.
    Date: 2018–03

This nep-for issue is ©2018 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.