nep-for New Economics Papers
on Forecasting
Issue of 2010‒10‒30
thirteen papers chosen by
Rob J Hyndman
Monash University

  1. Real-time forecast averaging with ALFRED By Chanont Banternghansa; Michael W. McCracken
  2. The information content of high-frequency data for estimating equity return models and forecasting risk By Dobrislav Dobrev; Pawel Szerszen
  3. Equity premium predictions with adaptive macro indexes By Jennie Bai
  4. A monthly consumption indicator for Germany based on internet search query data By Torsten Schmidt; Simeon Vosen
  5. Affecting Policy by Manipulating Prediction Markets: Experimental Evidence By Cary Deck; Shengle Lin; David Porter
  6. Currency Carry Trades By Travis J. Berge; Òscar Jordà; Alan M. Taylor
  7. Demographics and the Econometrics of the Term Structure of Stock Market Risk By Carlo A. Favero; Andrea Tamoni
  8. Asymmetries, breaks, and long-range dependence: An estimation framework for daily realized volatility By ERIC HILLEBRAND; MArcelo Cunha Medeiros
  9. Density-Conditional Forecasts in Dynamic Multivariate Models By Andersson, Michael K.; Palmqvist, Stefan; Waggoner, Daniel F.
  10. Modeling And Forecasting Imported Japanese Parts Content Of US Transplants: An Error Correction And State Space Approach By Cadogan, Godfrey
  11. Measures of Predictive Success for Rating Functions By Sebastian Ostrowski; Peter Reichling
  12. Insuring against loss of evidence in game-theoretic probability By A. Philip Dawid; Steven de Rooij; Glenn Shafer; Alexander Shen; Nikolai Vereshchagin; Vladimir Vovk
  13. A Perspective on Predicting Currency Crises By Juan Yepez; Robert P. Flood; Nancy P. Marion

  1. By: Chanont Banternghansa; Michael W. McCracken
    Abstract: This paper presents empirical evidence on the efficacy of forecast averaging using the ALFRED real-time database. We consider averages taken over a variety of different bivariate VAR models that are distinguished from one another based upon at least one of the following: which variables are used as predictors, the number of lags, using all available data or data after the Great Moderation, the observation window used to estimate the model parameters and construct averaging weights, and for forecast horizons greater than one, whether or not iterated- or direct-multistep methods are used. A variety of averaging methods are considered. Our results indicate that the benefits to model averaging relative to BIC-based model selection are highly dependent upon the class of models being averaged over. We provide a novel decomposition of the forecast improvements that allows us to determine which types of averaging methods and models were most (and least) useful in the averaging process.
    Keywords: Economic forecasting ; Real-time data
    Date: 2010
  2. By: Dobrislav Dobrev; Pawel Szerszen
    Abstract: We demonstrate that the parameters controlling skewness and kurtosis in popular equity return models estimated at daily frequency can be obtained almost as precisely as if volatility is observable by simply incorporating the strong information content of realized volatility measures extracted from high-frequency data. For this purpose, we introduce asymptotically exact volatility measurement equations in state space form and propose a Bayesian estimation approach. Our highly efficient estimates lead in turn to substantial gains for forecasting various risk measures at horizons ranging from a few days to a few months ahead when taking also into account parameter uncertainty. As a practical rule of thumb, we find that two years of high frequency data often suffice to obtain the same level of precision as twenty years of daily data, thereby making our approach particularly useful in finance applications where only short data samples are available or economically meaningful to use. Moreover, we find that compared to model inference without high-frequency data, our approach largely eliminates underestimation of risk during bad times or overestimation of risk during good times. We assess the attainable improvements in VaR forecast accuracy on simulated data and provide an empirical illustration on stock returns during the financial crisis of 2007-2008.
    Date: 2010
  3. By: Jennie Bai
    Abstract: Fundamental economic conditions are crucial determinants of equity premia. However, commonly used predictors do not adequately capture the changing nature of economic conditions and hence have limited power in forecasting equity returns. To address the inadequacy, this paper constructs macro indexes from large data sets and adaptively chooses optimal indexes to predict stock returns. I find that adaptive macro indexes explain a substantial fraction of the short-term variation in future stock returns and have more forecasting power than both the historical average of stock returns and commonly used predictors. The forecasting power exhibits a strong cyclical pattern, implying the ability of adaptive macro indexes to capture time-varying economic conditions. This finding highlights the importance of using dynamically measured economic conditions to investigate empirical linkages between the equity premium and macroeconomic fundamentals.
    Keywords: Stocks - Rate of return ; Forecasting ; Macroeconomics ; Economic indicators
    Date: 2010
  4. By: Torsten Schmidt; Simeon Vosen
    Abstract: In this study we introduce a new monthly indicator for private consumption in Germany based on search query time series provided by Google Trends. The indicator is based on unobserved factors extracted from a set of consumption-related search categories of the Google Trends application Insights for Search. The predictive performance of the indicator is assessed in real time relative to the European Commission’s consumer confidence indicator and the European Commission’s retail trade confidence indicator. In out-of-sample nowcasting experiments the Google indicator outperformed the surveybased indicators. In comparison to the other indicators, the new indicator also provided substantial predictive information on consumption beyond that already captured in other macroeconomic variables.
    Keywords: Google Trends, Private Consumption, Forecasting, Consumer Sentiment Indicator
    JEL: C53 E21 E27
    Date: 2010–10
  5. By: Cary Deck (University of Arkansas and Economic Science Institute); Shengle Lin (Economic Science Institute, Chapman University); David Porter (Economic Science Institute, Chapman University)
    Abstract: Documented results indicate prediction markets effectively aggregate information and form accurate predictions. This has led to a proliferation of markets predicting everything from the results of elections to a company’s sales to movie box office receipts. Recent research suggests prediction markets are robust to manipulation attacks and resulting market outcomes improve forecast accuracy. However, we present evidence from the lab indicating that well funded, single minded manipulators can in fact destroy a prediction market’s ability to aggregate information. Our results clearly indicate that the usefulness of prediction markets as inputs to decision making may be limited.
    Keywords: Information Aggregation, Prediction Markets, Manipulation
    JEL: C9 D8 G1
    Date: 2010
  6. By: Travis J. Berge; Òscar Jordà; Alan M. Taylor
    Abstract: A wave of recent research has studied the predictability of foreign currency returns. A wide variety of forecasting structures have been proposed, including signals such as carry, value, momentum, and the forward curve. Some of these have been explored individually, and others have been used in combination. In this paper we use new econometric tools for binary classification problems to evaluate the merits of a general model encompassing all these signals. We find very strong evidence of forecastability using the full set of signals, both in sample and out-of-sample. This holds true for both an unweighted directional forecast and one weighted by returns. Our preferred model generates economically meaningful returns on a portfolio of nine major currencies versus the U.S. dollar, with favorable Sharpe and skewness characteristics. We also find no relationship between our returns and a conventional set of so-called risk factors.
    JEL: C44 F31 F37 G14 G15
    Date: 2010–10
  7. By: Carlo A. Favero; Andrea Tamoni
    Abstract: The term structure of the stock market risk, defined as the per period conditional variance of cumulative returns, is measured in the strategic asset allocation literature (e.g. Campbell and Viceira (2002), (2005)) via multi-step ahead predictions from a VAR model of the joint process for one-period returns and their predictor, the dividend-price ratio. In this paper we modify the dynamic dividend growth model to allow for a time varying linearization point driven by the age structure of population. This specification leads to a decomposition of the dividend-price prices into an high volatility little persistence “noise” component, and a low volatility high persistence “information” component. The dividend-price ratio is mean reverting toward the time-varying mean and its deviations from it have a predicting power for returns that increases with the horizon. As a result of these two effects, the forward solution of the model delivers a negative sloping term structure of stock market risk. Direct regressions of returns at different horizons on the relevant predictors are much better suited to capture this feature than VAR based multi-period iterated forecasts. This evidence is very little affected by parameters’ uncertainty and is robust to the existence of "imperfect predictiors", as a parsimoniuos parameterization is very precisely estimated and no-projections for future variables are needed in the direct regression approach.
    Date: 2010
    Abstract: We study the simultaneous occurrence of long memory and nonlinear effects, such as structural breaks and thresholds, in autoregressive moving average (ARMA) time series models and apply our modeling framework to series of daily realized volatility. Asymptotic theory for the quasi-maximum likelihood estimator is developed and a sequence of model specification tests is described. Our framework allows for general nonlinear functions, including smoothly changing intercepts. The theoretical results in the paper can be applied to any series with long memory and nonlinearity. We apply the methodology to realized volatility of individual stocks of the Dow Jones Industrial Average during the period 1995 to 2005. We find strong evidence of nonlinear effects and explore different specifications of the model framework. A forecasting exercise demonstrates that allowing for nonlinearities in long memory models yields significant performance gains.
    Keywords: Realized volatility, structural breaks, smooth transitions, nonlinear models, long memory, persistence.
    Date: 2010–10
  9. By: Andersson, Michael K. (Monetary Policy Department, Central Bank of Sweden); Palmqvist, Stefan (Monetary Policy Department, Central Bank of Sweden); Waggoner, Daniel F. (Research Department)
    Abstract: When generating conditional forecasts in dynamic models it is common to impose the conditions as restrictions on future structural shocks. However, these conditional forecasts often ignore that there may be uncertainty about the future development of the restricted variables. Our paper therefore proposes a generalization such that the conditions can be given as the full distribution of the restricted variables. We demonstrate, in two empirical applications, that ignoring the uncertainty about the conditions implies that the distributions of the unrestricted variables are too narrow.
    Keywords: Central Bank; Market Expectation; Restrictions; Uncertainty
    JEL: C53 E37 E52
    Date: 2010–09–01
  10. By: Cadogan, Godfrey
    Abstract: This paper provides a sectoral examination of the impact of trade policies and custom valuation procedures on estimating time varying import content of Japanese transplant automobiles. Using monthly data from 1985 to 1992, we introduce an error correction model (ECM) and a state space VAR model to purify trade data of measurement errors induced by unobserveable prices and customs valuation procedures. Data show that US import of Japanese auto parts are elastic to the fleet of active Japanese automobiles in the U.S., inelastic to transplant production and that disequilibrium adjustments relative to transplant production are corrected in one period. Further, changes in imports are responsive to the cyclical behavior of Big 3 production and the debt burden of automobile consumers. Moreover, we find that productivity trends in the automotive industry are not a significant determinant of imported parts. The model predicts that Japanese manufacturers will shift more production to the US in response to yen appreciation against the dollar. We show that whereas import content decreased following the Fair Trade in Parts Act and the Omnibus Trade and Competitiveness Act of 1988, it increased shortly thereafter and predictions are that it will continue to increase. Therefore the empirical evidence suggests that direct trade policies designed to reduce import content and increase domestic sourcing of auto parts are not effective in the long run.
    Keywords: measurement error; error correction; state space forecasts; time varying import content
    JEL: C51 F14 F47 C13 C53 F17 C30 F21
    Date: 2010–09
  11. By: Sebastian Ostrowski (Faculty of Economics and Management, Otto-von-Guericke University Magdeburg); Peter Reichling (Faculty of Economics and Management, Otto-von-Guericke University Magdeburg)
    Abstract: Aim of our paper is to develop an adequate measure of predictive success and accuracy of rating functions. At first, we show that the common measures of rating accuracy, i.e. area under curve and accuracy ratio, respectively, lack of informative value of single rating classes. Selten (1991) builds up an axiomatic framework for measures of predictive success. Therefore, we introduce a measure for rating functions that fulfills the axioms proposed by Selten (1991). Furthermore, an empirical investigation analyzes predictive power and accuracy of Standard & Poor's and Moody's ratings, and compares the rankings according to area under curve and our measure.
    Keywords: Accuracy Measure, Rating Functions, Predictive Success, Discriminative Power
    JEL: C52 G21 G33
    Date: 2010–08
  12. By: A. Philip Dawid; Steven de Rooij; Glenn Shafer; Alexander Shen; Nikolai Vereshchagin; Vladimir Vovk
    Abstract: We consider the game-theoretic scenario of testing the performance of Forecaster by Sceptic who gambles against the forecasts. Sceptic's current capital is interpreted as the amount of evidence he has found against Forecaster. Reporting the maximum of Sceptic's capital so far exaggerates the evidence. We characterize the set of all increasing functions that remove the exaggeration. This result can be used for insuring against loss of evidence.
    Date: 2010–05
  13. By: Juan Yepez; Robert P. Flood; Nancy P. Marion
    Abstract: Currency crises are difficult to predict. It could be that we are choosing the wrong variables or using the wrong models or adopting measurement techniques not up to the task. We set up a Monte Carlo experiment designed to evaluate the measurement techniques. In our study, the methods are given the right fundamentals and the right models and are evaluated on how closely the estimated predictions match the objectively correct predictions. We find that all methods do reasonably well when fundamentals are explosive and all do badly when fundamentals are merely highly volatile.
    Date: 2010–10–13

This nep-for issue is ©2010 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.