nep-for New Economics Papers
on Forecasting
Issue of 2015‒11‒15
nine papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting Russian Macroeconomic Indicators with BVAR By Boris B. Demeshev; Oxana A. Malakhovskaya
  2. Oil Price Forecasts for the Long-Term: Expert Outlooks, Models, or Both? By Jean-Thomas Bernard; Lynda Khalaf; Maral Kichian; Clement Yelou
  3. Can consumer confidence provide independent information on consumption spending? By Antonello D’Agostino
  4. Forecasting commodity currencies: the role of fundamentals with short-lived predictive content By Claudia Foroni; Francesco Ravazzolo; Pinho J. Ribeiro
  5. Large Bayesian VARs: A flexible Kronecker error covariance structure By Joshua C.C. Chan
  6. A Composite Likelihood Framework for Analyzing Singular DSGE Models By Zhongjun Qu
  7. Estimating Interest Rate Setting Behavior in Korea: A Constrained Ordered Choices Model Approach By Hyeongwoo Kim; Wen Shi; Kwang-Myoung Hwang
  8. Working Paper – WP/14/04- A medium-sized open economy DSGE model of South Africa By Stan du Plessis; Ben Smit; Rudi Steinbach
  9. [JOB MARKET PAPER] Sophisticated Trading and Market Efficiency: Evidence from Macroeconomic News Announcements By John C. Zhou

  1. By: Boris B. Demeshev (National Research University Higher School of Economics); Oxana A. Malakhovskaya (National Research University Higher School of Economics)
    Abstract: This paper evaluates the forecast performance of Bayesian vector autoregressions (BVARs) on Russian data. We estimate BVARs of different sizes and compare the accuracy of their out-ofsample forecasts with those obtained with unrestricted vector autoregressions and random walk with drift. We show that many Russian macroeconomic indicators can be forecast by BVARs more accurately than by competing models. However, contrary to several other studies, we do not confirm that the relative forecast error monotonically decreases with increasing the crosssectional dimension of the sample. In half of those cases where a BVAR appears to be the most accurate model, a small-dimensional BVAR outperforms its high-dimensional counterpart.
    Keywords: VAR, BVAR, forecasting, Bayesian estimation
    JEL: C11 C13 C53
    Date: 2015
  2. By: Jean-Thomas Bernard (Department of Economics, University of Ottawa); Lynda Khalaf (Carleton University); Maral Kichian (University of Ottawa); Clement Yelou (Laval University)
    Abstract: Expert outlooks on the future path of oil prices are often relied on by industry participants and policymaking bodies for their forecasting needs. Yet little attention has been paid to the extent to which these are accurate. Using the regular publications by the Energy Information Administration (EIA), we examine the accuracy of annual recursive oil price forecasts generated by the National Energy Modeling System model of the Agency for forecast horizons of up to 15 years. Our results reveal that the EIA model is quite successful at beating the benchmark random walk model, but only at either end of the forecast horizons. We also show that, for the longer horizons, simple econometric forecasting models often produce similar if not better accuracy than the EIA model. Among these, time-varying specifications generally also exhibit stability in their forecast performance. Finally, while combining forecasts does not change the overall patterns, some additional accuracy gains are obtained at intermediate horizons, and in some cases forecast performance stability is also achieved.
    Keywords: Oil price, expert outlooks, long run forecasting, forecast combinations
    JEL: Q47 C20
    Date: 2015
  3. By: Antonello D’Agostino (ESM)
    Abstract: This paper investigates how well consumer confidence predicts households future consumption expenditure. Our findings document considerable variety in the degree to which confidence measures accurately forecast consumption across selected euro area countries and periods. First, we explore the leading role of consumer confidence in forecasting consumption growth. We find that the consumer confidence index improves forecasts of household consumption expenditure appreciably during times of financial distress, especially in Italy and Portugal. Further, we show that the financial sub-index of consumer confidence provides more nuanced information than the aggregate index. Indeed, over the past few years, expectations about future personal financial situations proved particularly helpful in forecasting total consumption expenditure in France, Italy and Portugal. For Germany, in contrast, no measures of confidence provide information beyond what is supplied by other economic indicators for forecasting household consumption. Finally, we advance some evidence to support the idea that changes in consumer confidence are an independent driver of economic fluctuations.
    Keywords: Expectations; Survey Data; Consumption Forecast; Confidence Shocks; Economic Fluctuations
    JEL: C32 E24 E32
  4. By: Claudia Foroni (Norges Bank); Francesco Ravazzolo (Norges Bank and Centre for Applied Macro and Petroleum Economics at BI Norwegian Business School); Pinho J. Ribeiro (University of Glasgow, Adam Smith Business School)
    Abstract: Recent evidence highlights that commodity price changes exhibit a short-lived, yet robust contemporaneous effect on commodity currencies, which is mainly detectable in daily-frequency data. We use MIDAS models in a Bayesian setting to include mixed-frequency dynamics while accounting for time-variation in predictive ability. Using the random walk Metropolis-Hastings technique as a new tool to estimate our class of MIDAS regressions, we find that for most of the commodity currencies in our sample exploiting this short-lived relationship yields to statistically more precise out-of-sample exchange rate point and density forecasts relative to the no-change benchmark. Further, the usual low-frequency predictors, such as money supplies and interest rates differentials, typically receive little support from the data at monthly forecasting horizons. In contrast, models featuring daily commodity prices are highly likely.
    Keywords: Exchange rate point and density forecasting; Commodity prices; MIDAS model; Bayesian model averaging; Metropolis-Hastings algorithm
    JEL: C53 F37
    Date: 2015–10–30
  5. By: Joshua C.C. Chan
    Abstract: We introduce a class of large Bayesian vector autoregressions (BVARs) that allows for non-Gaussian, heteroscedastic and serially dependent innovations. To make estimation computationally tractable, we exploit a certain Kronecker structure of the likelihood implied by this class of models. We propose a unified approach for estimating these models using Markov chain Monte Carlo (MCMC) methods. In an application that involves 20 macroeconomic variables, we find that these BVARs with more flexible covariance structures outperform the standard variant with independent, homoscedastic Gaussian innovations in both in-sample model-fit and out-of-sample forecast performance.
    Keywords: stochastic volatility, non-Gaussian, ARMA, forecasting
    JEL: C11 C51 C53
    Date: 2015–11
  6. By: Zhongjun Qu (Boston University)
    Abstract: This paper builds upon the composite likelihood concept of Lindsay (1988) to develop a framework for parameter identification, estimation, inference and forecasting in DSGE models allowing for stochastic singularity. The framework consists of the following four components. First, it provides a necessary and sufficient condition for parameter identification, where the identifying information is provided by the first and second order properties of the nonsingular submodels. Second, it provides an MCMC based procedure for parameter estimation. Third, it delivers confidence sets for the structural parameters and the impulse responses that allow for model misspecification. Fourth, it generates forecasts for all the observed endogenous variables, irrespective of the number of shocks in the model. The framework encompasses the conventional likelihood analysis as a special case when the model is nonsingular. Importantly, it enables the researcher to start with a basic model and then gradually incorporate more shocks and other features, meanwhile confronting all the models with the data to assess their implications. The methodology is illustrated using both small and medium scale DSGE models. These models have numbers of shocks ranging between one and seven.
    Keywords: business cycle, dynamic stochastic general equilibrium models, identification, impulse response, MCMC, stochastic singularity
    JEL: C13 C32 C51 E1
    Date: 2015–06
  7. By: Hyeongwoo Kim; Wen Shi; Kwang-Myoung Hwang
    Abstract: We study the Bank of Korea’s interest rate setting behavior using an array of constrained ordered choices models, where the Monetary Policy Committee revises the target policy interest rate only when the current market interest rate deviates from the optimal rate by more than certain threshold values. Our models explain changes in the monetary policy stance well for the monthly frequency Korean data since January 2000. We find important roles for the output gap and the foreign exchange rate in understanding the Bank of Korea’s rate decision-making process. We also implement out-of-sample forecast exercises with September 2008 (Lehman Brothers Bankruptcy) for a split point. We demonstrate that out-of-sample predictability improves greatly for the rate cut and the rate hike decisions using standard error adjusted inaction bands.
    Keywords: Monetary Policy; Bank of Korea; Probit Model; Robit Model; Logit Model; Target RP Rate; Interbank Call Rate; Taylor Rule
    JEL: C51 C52 E52 E58
    Date: 2015–11
  8. By: Stan du Plessis; Ben Smit; Rudi Steinbach
    Abstract: In this paper a dynamic stochastic general equilibrium (DSGE) model is specified for the South African economy. Nominal and real frictions help to make the model estimable, and is then estimated on South African and global data using Bayesian techniques. The empirical fit of the model is validated through a forecast comparison with private sector consensus forecasts. The model is found to outperform the inflation forecasts of private sector economists.
    Date: 2014–07–11
  9. By: John C. Zhou
    Abstract: This paper studies how the views of sophisticated traders are impounded into stocks and bonds around macroeconomic news announcements. I find evidence that sophisticated traders trade on predictions of macroeconomic news reports before announcements and obtain their informational advantage using public information. Specifically, consensus forecasts of upcoming data releases suffer from anchoring bias and overweight past data releases. By correcting this bias, sophisticated traders can predict news reports. The results suggest that stock and bond markets are inefficient in this setting. Over time, there is a late trading puzzle: sophisticated traders can predict news reports days before announcements but appear to trade these predictions into stock and bond prices just hours before announcements. Across assets there is a related puzzle: the predictable component of news reports is eventually fully impounded into bonds but only partially impounded into stocks. Stocks but not bonds react to announcements of the predictable component and display return momentum. Using a model, I argue that market inefficiency can arise when unsophisticated traders neglect public information that predicts news reports, and risk management concerns deter sophisticated traders from acting on their informational edge. Trading earlier and trading riskier assets such as stocks exposes sophisticated traders to greater risk. As a result, sophisticated traders wait to trade and trade safer assets such as bonds.
    Date: 2015–11

This nep-for issue is ©2015 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.