nep-for New Economics Papers
on Forecasting
Issue of 2006‒04‒08
twelve papers chosen by
Rob J Hyndman
Monash University

  1. Cointegration, Integration, and Long-Term Forcasting By Hiroaki Chigira; Taku Yamamoto
  2. Forecasting Substantial Data Revisions in the Presence of Model Uncertainty By Troy Matheson
  3. Marketwide Private Information in Stocks: Forecasting Currency Returns By Albuquerque, Rui; de Francisco, Eva; Marques, Luis
  4. Monetary Policy Transparency in the UK:The Impact of Independence and Inflation Targeting By Iris Biefang-Frisancho Mariscal; Peter Howells
  5. Measuring the Volatility in U.S. Treasury Benchmarks and Debt Instruments By Suhejla Hoiti; Esfandiar Maasoumi; Michael McAleer; Daniel Slottje
  6. The Volatility of Realized Volatility By Fulvio Corsi; Uta Kretschmer; Stefan Mittnik; Christian Pigorsch
  7. Competing Approaches to Forecasting Elections: Economic Models, Opinion Polling and Prediction Markets By Leigh, Andrew; Wolfers, Justin
  8. Prediction Markets in Theory and Practice By Wolfers, Justin; Zitzewitz, Eric
  9. Five Open Questions About Prediction Markets By Wolfers, Justin; Zitzewitz, Eric
  10. Phillips curve forecasting in a small open economy By Anthony Garratt; Gary Koop; Shaun P. Vahey
  11. Modeling Travel Demand in a Metropolitan City: Case Study of Bangalore, India By Pangotra Prem; Sharma Somesh
  12. Measuring Expectations By Kjellberg, David

  1. By: Hiroaki Chigira; Taku Yamamoto
    Abstract: It is widely believed that taking cointegration and integration into consideration is useful in constructing long-term forecasts for cointegrated processes. This paper shows that imposing neither cointegration nor integration leads to superior long-term forecasts.
    Keywords: Forecasting, Cointegration, Integration
    JEL: C12 C32
    Date: 2006–03
  2. By: Troy Matheson (Reserve Bank of New Zealand)
    Abstract: Stock and Watson (1999) show that the Phillips curve is a good forecasting tool in the United States. We assess whether this good performance extends to two small open economies, with relatively large tradable sectors. Using data for Australia and New Zealand, we find that the open economy Phillips curve performs poorly relative to a univariate autoregressive benchmark. However, its performance improves markedly when sectoral Phillips curves are used which model the tradable and non-tradable sectors separately. Combining forecasts from these sectoral models is much better than obtaining forecasts from a Phillips curve estimated on aggregate data. We also find that a diffusion index that combines a large number of indicators of real economic activity provides better forecasts of non-tradable inflation than more conventional measures of real demand, thus supporting Stock and Watson's (1999) findings for the United States.
    JEL: C53 E31
    Date: 2006–02
  3. By: Albuquerque, Rui; de Francisco, Eva; Marques, Luis
    Abstract: We present a model of equity trading with informed and uninformed investors where informed investors act upon firm-specific private information and marketwide private information. The model is used to structurally identify the component of order flow that is due to marketwide private information. Trades driven by marketwide private information display very little or no correlation with the first principal component of order flow. This finding implies that a simple statistical factor is a poor measure of marketwide private information. Moreover, the model suggests that the previously documented comovement in order flow captures mostly common variation in liquidity trades. We find that marketwide private information obtained from equity market data forecasts industry stock returns and foreign exchange returns consistent with Evans and Lyons' (2004a) model of exchange rate determination.
    Keywords: currency returns; equity returns; firm-specific private information; marketwide private information; order flow; principal components
    JEL: F31 G11 G14
    Date: 2006–03
  4. By: Iris Biefang-Frisancho Mariscal; Peter Howells (School of Economics, University of the West of England)
    Abstract: There is a widespread belief that the transparency of UK monetary policy has increased substantially as a result of the introduction of inflation targeting in 1992 and a number of procedural and institutional reforms which accompanied and followed it. In this paper, we use money market responses (and other data) to test the possibility that improved anticipation of policy moves may be the result of developments other than the institutional reforms popularly cited. We find overwhelming evidence that the switch to inflation targeting itself significantly reduced monetary policy surprises, while subsequent reforms have contributed little. Where we advance substantially on earlier work is to look at the cross-sectional dispersion of agents’ anticipation. If the benefit of transparency is the elimination of policy surprise, there is little benefit if the averagely correct anticipations of agents conceal a wide dispersion of view. The most striking feature is the general decline in cross-sectional one year-ahead forecast uncertainty of the interbank rate. So, even though we do not find that agents on average have improved monetary policy anticipation since 1997, we do find that they have become more unanimous about forecasting future money market rates. However, further testing reveals that it is a simultaneous fall in the dispersion of inflation rate forecasts that explains the increased consensus on interest rates, rather than institutional reforms in 1997 and later.
    Keywords: Monetary Policy; transparency; independence; inflation targeting
    JEL: E58
    Date: 2006–01
  5. By: Suhejla Hoiti; Esfandiar Maasoumi; Michael McAleer; Daniel Slottje
    Abstract: As U.S. Treasury securities carry the full faith and credit of the U.S. government, they are free of default risk. Thus, their yields are risk-free rates of return, which allows the most recently issued U.S. Treasury securities to be used as a benchmark to price other fixedincome instruments. This paper analyzes the time series properties of interest rates on U.S. Treasury benchmarks and related debt instruments by modelling the conditional mean and conditional volatility for weekly yields on 12 Treasury Bills and other debt instruments for the period 8 January 1982 to 20 August 2004. The conditional correlations between all pairs of debt instruments are also calculated. These estimates are of interest as they enable an assessment of the implications of modelling conditional volatility on forecasting performance. The estimated conditional correlation coefficients indicate whether there is specialization, diversification or independence in the debt instrument shocks. Constant conditional correlation estimates of the standardized shocks indicate that the shocks to the first differences in the debt instrument yields are generally high and always positively correlated. In general, the primary purpose in holding a portfolio of Treasury Bills and other debt instruments should be to specialize on instruments that provide the largest returns. Tests for Stochastic Dominance are consistent with these findings, but find somewhat surprising rankings between debt instruments with implications for portfolio composition. 30 year treasuries, Aaa bonds and mortgages tend to dominate other instruments, at least to the second order.
    Keywords: Treasury bills, debt instruments, risk, conditional volatility, conditional correlation, asymmetry, specialization, diversification, independence, forecasting.
    Date: 2005–10
  6. By: Fulvio Corsi (University of Lugano); Uta Kretschmer (University of Bonn, Germany); Stefan Mittnik (University of Munich); Christian Pigorsch (University of Munich)
    Abstract: Using unobservable conditional variance as measure, latent–variable approaches, such as GARCH and stochastic–volatility models, have traditionally been dominating the empirical finance literature. In recent years, with the availability of high–frequency financial market data modeling realized volatility has become a new and innovative research direction. By constructing “observable” or realized volatility series from intraday transaction data, the use of standard time series models, such as ARFIMA models, have become a promising strategy for modeling and predicting (daily) volatility. In this paper, we show that the residuals of the commonly used time–series models for realized volatility exhibit non–Gaussianity and volatility clustering. We propose extensions to explicitly account for these properties and assess their relevance when modeling and forecasting realized volatility. In an empirical application for S&P500 index futures we show that allowing for time–varying volatility of realized volatility leads to a substantial improvement of the model’s fit as well as predictive performance. Furthermore, the distributional assumption for residuals plays a crucial role in density forecasting.
    Keywords: Finance, Realized Volatility, Realized Quarticity, GARCH, Normal Inverse Gaussian Distribution, Density Forecasting
    JEL: C22 C51 C52 C53
    Date: 2005–11–28
  7. By: Leigh, Andrew; Wolfers, Justin
    Abstract: We review the efficacy of three approaches to forecasting elections: econometric models that project outcomes on the basis of the state of the economy; public opinion polls; and election betting (prediction markets). We assess the efficacy of each in light of the 2004 Australian election. This election is particularly interesting both because of innovations in each forecasting technology, and also because the increased majority achieved by the Coalition surprised most pundits. While the evidence for economic voting has historically been weak for Australia, the 2004 election suggests an increasingly important role for these models. The performance of polls was quite uneven, and predictions both across pollsters, and through time, vary too much to be particularly useful. Betting markets provide an interesting contrast, and a slew of data from various betting agencies suggests a more reasonable degree of volatility, and useful forecasting performance both throughout the election cycle and across individual electorates.
    Keywords: elections; macroeconomic voting; opinion polling; prediction markets; voting
    JEL: D72 D84
    Date: 2006–03
  8. By: Wolfers, Justin; Zitzewitz, Eric
    Abstract: Prediction Markets, sometimes referred to as 'information markets', 'idea futures' or 'event futures', are markets where participants trade contracts whose payoffs are tied to a future event, thereby yielding prices that can be interpreted as market-aggregated forecasts. This article summarizes the recent literature on prediction markets, highlighting both theoretical contributions that emphasize the possibility that these markets efficiently aggregate disperse information, and the lessons from empirical applications which show that market-generated forecasts typically outperform most moderately sophisticated benchmarks. Along the way, we highlight areas ripe for future research.
    Keywords: event futures; forecasting; futures; information aggregation; information markets; prediction markets
    JEL: C53 D8 G14
    Date: 2006–03
  9. By: Wolfers, Justin; Zitzewitz, Eric
    Abstract: Interest in prediction markets has increased in the last decade, driven in part by the hope that these markets will prove to be valuable tools in forecasting, decision-making and risk management - in both the public and private sectors. This paper outlines five open questions in the literature, and we argue that resolving these questions is crucial to determining whether current optimism about prediction markets will be realized.
    Keywords: event futures; information markets; instrumental variables; market manipulation; prediction IV; prediction markets
    JEL: C9 D7 D8 G1 M2
    Date: 2006–03
  10. By: Anthony Garratt; Gary Koop; Shaun P. Vahey (Reserve Bank of New Zealand)
    Abstract: A recent revision to the preliminary measurement of GDP(E) growth for 2003Q2 caused considerable press attention, provoked a public enquiry and prompted a number of reforms to UK statistical reporting procedures. In this paper, we compute the probability of "substantial revisions" that are greater (in absolute value) than the controversial 2003 revision. The pre-dictive densities are derived from Bayesian model averaging over a wide set of forecasting models including linear, structural break and regime-switching models with and without heteroskedasticity. Ignoring the nonlinearities and model uncertainty yields misleading predictives and obscures the improvement in the quality of preliminary UK macroeconomic measurements relative to the early 1990s.
    JEL: C11 C32 C53
    Date: 2006–02
  11. By: Pangotra Prem; Sharma Somesh
    Abstract: Increasing urbanization, population growth and rising incomes have led to rapid growth of travel demand in Indian cities. The paper provides a modeling approach for forecasting urban travel demand and assessing public transport options for large metropolitan cities. A travel characteristics model is used to forecast the pattern of travel demand in Bangalore city up to the year 2014. The paper examines the scope of a public bus transport service and a mass rapid transit system for meeting the projected travel demand and thereby curtailing the growth of personal vehicles in the city.
    Date: 2006–03–28
  12. By: Kjellberg, David (Department of Economics)
    Abstract: To evaluate measures of expectations I examine and compare some of the most common methods for capturing expectations: the futures method which utilizes financial market prices, the VAR forecast method, and the survey method. I study average expectations on the Federal funds rate target, and the main findings can be summarized as follows: i) the survey measure and the futures measure are highly correlated; the correlation coefficient is 0.81 which indicates that the measures capture the same phenomenon, ii) the survey measure consistently overestimates the realized changes in the interest rate, iii) the VAR forecast method shows little resemblance with the other methods.
    Keywords: Interest rates; expectations; futures; VAR forecasts; survey data
    JEL: E43 E44 E47
    Date: 2006–02

This nep-for issue is ©2006 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.