nep-for New Economics Papers
on Forecasting
Issue of 2015‒01‒26
nine papers chosen by
Rob J Hyndman
Monash University

  1. Speculative behaviour and oil price predictability By Ekaterini Panopoulou; Theologos Pantelidis
  2. Assessing the Short-term Forecasting Power of Confidence Indices By Euler Pereira G. de Mello; Francisco Marcos R. Figueiredo
  3. Can Oil Prices Forecast Exchange Rates? By Domenico Ferraro; Ken Rogoff; Barbara Rossi
  4. Censored spatial wind power prediction with random effects By Croonenbroeck, Carsten; Ambach, Daniel
  5. Predicting Economic Activity via Eurozone Yield Spreads: Impact of Credit Risk By Schock, Matthias
  6. Combined Density Nowcasting in an uncertain economic environment By Knut Are Aastveit; Francesco Ravazzolo; Herman K. van Dijk
  7. Inferring inflation expectations from fixed-event forecasts By Winkelried, Diego
  8. What predicts financial (in)stability? A Bayesian approach By Eidenberger, Judith; Neudorfer, Benjamin; Sigmund, Michael; Stein, Ingrid
  9. Inflation Uncertainty and Disagreement in Bond Risk Premia By D'Amico, Stefania; Orphanides, Athanasios

  1. By: Ekaterini Panopoulou (Kent Business School, University of Kent); Theologos Pantelidis (Department of Economics, University of Macedonia)
    Abstract: We develop two- and three-state regime switching models and test their forecasting ability for oil prices. We use the deviations of market oil price from fundamental values as the main explanatory variable in our models, while additional potential predictors enrich our specification. Our findings suggest that the regime-switching models are, in general, more accurate than the Random Walk model in terms of both statistical and economic evaluation criteria for oil price forecasts.
    Keywords: Oil price; Regime Switching; Forecasting; Deviations from fundamentals.
    JEL: C5 Q4
    Date: 2014–12
  2. By: Euler Pereira G. de Mello; Francisco Marcos R. Figueiredo
    Abstract: This paper assesses the predictive power of the main confidence índices available in Brazil to forecast economic activity. More specifically, we consider a set of economic activity variables and, for each of those, compare the predictive power of a univariate autoregressive model to that of a similar model that includes confidence index. Preliminary results using the Diebold Mariano test suggest that the Industry Confidence Index (ICI) provides relevant information, for both present and the near future, on some economic activity variables of interest to the economic agents
    Date: 2014–12
  3. By: Domenico Ferraro; Ken Rogoff; Barbara Rossi
    Abstract: We show the existence of a very short-term relationship at the daily frequency between changes in the price of a country's major commodity export price and changes in its nominal exchange rate. The relationship appears to be robust and to hold when we use contemporaneous (realized) commodity price changes in our regression. How- ever, when we use lagged commodity price changes, the predictive ability is ephemeral, mostly appearing after instabilities have been appropriately taken into account.
    Keywords: forecasting, oil prices, exchange rates
    JEL: F31 F37 C22 C53
    Date: 2015–01
  4. By: Croonenbroeck, Carsten; Ambach, Daniel
    Abstract: We investigate the importance of taking the spatial interaction of turbines inside a wind park into account. This article provides two tests that check for wake effects and thus, take spatial interdependence into account. Those effects are suspected to have a negative influence on wind power production. Thereafter, we introduce a new modeling approach that is based on the Generalized Wind Power Prediction Tool (GWPPT) and therefore respects both-sided censoring of the data. Furthermore, the new model takes a Spatial Lag Model (SLM) specification into account and allows for random effects in the panel data. Finally, we provide a short empirical study that compares the forecasting accuracy of our model to the established models WPPT, GWPPT, and the naïve persistence predictor. We show that our new model provides significantly better forecasts than the established models.
    Keywords: Spatial Lag Model,Censored,Regression,Wind Power,Forecasting,Random Effects
    JEL: C31 C34 E27 Q47
    Date: 2014
  5. By: Schock, Matthias
    Abstract: A "lost decade" for the Eurozone is looming on the horizon. Under these circumstances, stable indicators for future economic activity are especially valuable to decision makers. This paper examines the predictive power of the yield spread, one of the most reliable indicators for gross domestic product (GDP) growth. Despite the continuously high level of yield spreads, growth is sparse in the Eurozone. We find this to be caused by default risks, which are distorting the long-term interest rates of many Eurozone countries. Therefore, a new method of risk adjustment is introduced. We employ credit default swap (CDS) spreads on sovereign bonds, which provide a direct measure of credit risk. Incorporating those spreads significantly enhances the in- and out-of-sample predictive power of the yield-spread approach. Ordinary least squares (OLS) and fixed-effects models are used to forecast GDP growth in the Eurozone, and a probit model is used for recession prediction. The results show that the accuracy of predicting growth and recessions using the yield spread is high, provided that biases associated with Eurozone sovereign default risk are considered.
    Keywords: yield curve, CDS spreads, economic activity
    JEL: G1 E37 E43 E44
    Date: 2015–01
  6. By: Knut Are Aastveit (Norges Bank (Central Bank of Norway)); Francesco Ravazzolo (Norges Bank (Central Bank of Norway) and BI Norwegian Business School); Herman K. van Dijk (Erasmus University Rotterdam, VU University Amsterdam and Tinbergen Institute,)
    Abstract: We introduce a Combined Density Nowcasting (CDN) approach to Dynamic Factor Models (DFM) that in a coherent way accounts for time-varying uncertainty of several model and data features in order to provide more accurate and complete density nowcasts. The combination weights are latent random variables that depend on past nowcasting performance and other learning mechanisms. The combined density scheme is incorporated in a Bayesian Sequential Monte Carlo method which re-balances the set of nowcasted densities in each period using updated information on the time-varying weights. Experiments with simulated data show that CDN works particularly well in a situation of early data releases with relatively large data uncertainty and model incompleteness. Empirical results, based on US real-time data of 120 leading indicators, indicate that CDN gives more accurate density nowcasts of US GDP growth than a model selection strategy and other combination strategies throughout the quarter with relatively large gains for the two rst months of the quarter. CDN also provides informative signals on model incompleteness during recent recessions. Focusing on the tails, CDN delivers probabilities of negative growth, that provide good signals for calling recessions and ending economic slumps in real time.
    Keywords: Density forecast combination, Survey forecast, Bayesian filtering; Sequential Monte Carlo Nowcasting; Real-time data
    JEL: C11 C13 C32 C53 E37
    Date: 2014–12–04
  7. By: Winkelried, Diego (Universidad del Pacífico)
    Abstract: Often, expected inflation measured by surveys are available only as fixed-event forecasts. Even though these surveys do contain information of a complete term structure of expectations, direct inferences about them are troublesome. Records of a fixed-event forecast through time are associated with time-varying forecast horizons, and there is no straightforward way to interpolate such figures. This paper proposes an adaptation of the measurement model of Kozicki and Tinsley (2012) [“Effective use of survey information in estimating the evolution of expected inflation”, Journal of Money, Credit and Banking, 44(1), 145-169] to suit the intricacies of fixed-event data. Using the Latin American Consensus Forecasts, the model is estimated to study the behavior of inflation expectations in four inflation targeters (Chile, Colombia, Mexico and Peru). For these countries, the results suggest that the announcement of credible inflation targets has been instrumental in anchoring long-run expectations.
    Keywords: Survey expectations, fixed-event forecasts, Kalman filter, inflation targeting, Latin America
    JEL: C32 E37 E52
    Date: 2014–12
  8. By: Eidenberger, Judith; Neudorfer, Benjamin; Sigmund, Michael; Stein, Ingrid
    Abstract: This paper contributes to the literature on early warning indicators by applying a Bayesian model averaging approach. Our analysis, based on Austrian data, is carried out in two steps: First, we construct a quarterly financial stress index (AFSI) quantifying the level of stress in the Austrian financial system. Second, we examine the predictive power of various indicators, as measured by their ability to forecast the AFSI. Our approach allows us to investigate a large number of indicators. The results show that excessive credit growth and high returns of banks' stocks are the best early warning indicators. Unstable funding (as measured by the loan to deposit ratio) also has a high predictive power.
    Keywords: financial crisis,early warning indicators,government policy and regulation,financial stress index
    JEL: G01 G28
    Date: 2014
  9. By: D'Amico, Stefania (Federal Reserve Bank of Chicago); Orphanides, Athanasios (Federal Reserve Bank of Chicago)
    Abstract: This paper examines the relation between variations in perceived inflation uncertainty and bond premia. Using the subjective probability distributions available in the Survey of Professional Forecasters we construct a quarterly time series of average individual uncertainty about inflation forecasts since 1968. We show that this ex-ante measure of inflation uncertainty differs importantly from measures of disagreement regarding inflation forecasts and other proxies, such as model-based ex-post measures of macroeconomic risk. Inflation uncertainty is an important driver of bond premia, but the relation varies across inflation regimes. It is most important in the high-inflation regime early in the sample and the low-inflation regime over the last 15 years. Once the role of inflation uncertainty is accounted for, disagreement regarding inflation forecasts appears a much less important driver of bond premia.
    Keywords: Survey expectations; probabilistic forecasts; heterogeneity; inflation uncertainty; bond risk premia
    JEL: C53 E37 E44 E47 G12
    Date: 2014–01–11

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.