nep-for New Economics Papers
on Forecasting
Issue of 2010‒09‒03
twelve papers chosen by
Rob J Hyndman
Monash University

  1. Short-term load forecasting based on a semi-parametric additive model By Shu Fan; Rob Hyndman
  2. The Forecast Performance of Competing Implied Volatility Measures: The Case of Individual Stocks By Leonidas Tsiaras
  3. How helpful are spatial effects in forecasting the growth of Chinese provinces? By Girardin , Eric; Kholodilin, Konstantin A.
  4. Combining Non-Replicable Forecasts By Chang, C-L.; McAleer, M.J.; Franses, Ph.H.B.F.
  5. Analyzing and Forecasting Volatility Spillovers and Asymmetries in Major Crude Oil Spot, Forward and Futures Markets By Chia-Lin Chang; Michael McAleer; Roengchai Tansuchat
  6. The Role of Realized Ex-post Covariance Measures and Dynamic Model Choice on the Quality of Covariance Forecasts By Rasmus Tangsgaard Varneskov; Valeri Voev
  7. The Role of Dynamic Specification in Forecasting Volatility in the Presence of Jumps and Noisy High-Frequency Data By Rasmus Tangsgaard Varneskov
  8. Option Pricing with Asymmetric Heteroskedastic Normal Mixture Models By Jeroen V.K. Rombouts; Lars Stentoft
  9. Modelling and Forecasting UK Mortgage Arrears and Possessions By Janine Aron; John Muellbauer
  10. Dynamic Models of Exchange Rate Dependence Using Option Prices and Historical Returns By Leonidas Tsiaras
  11. Milking The Prices: The Role of Asymmetries in the Price Transmission Mechanism for Milk Products in Austria By Octavio Fernández-Amador; Josef Baumgartner; Jesús Crespo-Cuaresma
  12. Predictable return distributions By Thomas Q. Pedersen

  1. By: Shu Fan; Rob Hyndman
    Abstract: Short-term load forecasting is an essential instrument in power system planning, operation and control. Many operating decisions are based on load forecasts, such as dispatch scheduling of generating capacity, reliability analysis, and maintenance planning for the generators. Overestimation of electricity demand will cause a conservative operation, which leads to the start-up of too many units or excessive energy purchase, thereby supplying an unnecessary level of reserve. On the contrary, underestimation may result in a risky operation, with insufficient preparation of spinning reserve, causing the system to operate in a vulnerable region to the disturbance. In this paper, semi-parametric additive models are proposed to estimate the relationships between demand and the driver variables. Specifically, the inputs for these models are calendar variables, lagged actual demand observations and historical and forecast temperature traces for one or more sites in the target power system. In addition to point forecasts, prediction intervals are also estimated using a modified bootstrap method suitable for the complex seasonality seen in electricity demand data. The proposed methodology has been used to forecast the half-hourly electricity demand for up to seven days ahead for power systems in the Australian National Electricity Market. The performance of the methodology is validated via out-of-sample experiments with real data from the power system, as well as through on-site implementation by the system operator.
    Keywords: Short-term load forecasting, additive model, time series, forecast distribution
    JEL: C14 C15 C52 C53 L94
    Date: 2010–08–12
  2. By: Leonidas Tsiaras (Department of Business Studies, ASB, Aarhus University and CREATES)
    Abstract: This study examines the information content of alternative implied volatility measures for the 30 components of the Dow Jones Industrial Average Index from 1996 until 2007. Along with the popular Black-Scholes and \model-free" implied volatility expectations, the recently proposed corridor implied volatil- ity (CIV) measures are explored. For all pair-wise comparisons, it is found that a CIV measure that is closely related to the model-free implied volatility, nearly always delivers the most accurate forecasts for the majority of the firms. This finding remains consistent for different forecast horizons, volatility definitions, loss functions and forecast evaluation settings.
    Keywords: Model-Free Implied Volatility, Corridor Implied Volatility, Volatility Forecasting
    JEL: C22 C53 G13 G14
    Date: 2010–02–01
  3. By: Girardin , Eric (BOFIT); Kholodilin, Konstantin A. (BOFIT)
    Abstract: In this paper, we make multi-step forecasts of the annual growth rates of the real Gross Regional Product (GRP) for each of the 31 Chinese provinces simultaneously. Beside the usual panel data models, we use panel models that explicitly account for spatial dependence between the GRP growth rates. In addition, the possibility of spatial effects being different for different groups of provinces (Interior and Coast) is allowed for. We find that both pooling and accounting for spatial effects helps substantially improve the forecast performance compared to the benchmark models estimated for each of the provinces separately. It is also shown that the effect of accounting for spatial dependence is even more pronounced at longer forecasting horizons (the forecast accuracy gain as measured by the root mean squared forecast error is about 8% at the 1-year horizon and exceeds 25% at the 13- and 14-year horizon).
    Keywords: Chinese provinces; forecasting; dynamic panel model; spatial autocorrelation; group-specific spatial dependence
    JEL: C21 C23 C53
    Date: 2010–08–23
  4. By: Chang, C-L.; McAleer, M.J.; Franses, Ph.H.B.F.
    Abstract: Macro-economic forecasts are often based on the interaction between econometric models and experts. A forecast that is based only on an econometric model is replicable and may be unbiased, whereas a forecast that is not based only on an econometric model, but also incorporates an expert’s touch, is non-replicable and is typically biased. In this paper we propose a methodology to analyze the qualities of combined non-replicable forecasts. One part of the methodology seeks to retrieve a replicable component from the non-replicable forecasts, and compares this component against the actual data. A second part modifies the estimation routine due to the assumption that the difference between a replicable and a non-replicable forecast involves a measurement error. An empirical example to forecast economic fundamentals for Taiwan shows the relevance of the methodological approach.
    Keywords: combined forecasts;efficient estimation;generated regressors;replicable forecasts;non-replicable forecasts;expert’s intuition;C22;E27;E37
    Date: 2010–07–28
  5. By: Chia-Lin Chang (Department of Applied Economics, National Chung Hsing University); Michael McAleer (Erasmus University Rotterdam, Tinbergen Institute, The Netherlands, and Institute of Economic Research, Kyoto University); Roengchai Tansuchat (Faculty of Economics, Maejo University)
    Abstract: Crude oil price volatility has been analyzed extensively for organized spot, forward and futures markets for well over a decade, and is crucial for forecasting volatility and Value-at- Risk (VaR). There are four major benchmarks in the international oil market, namely West Texas Intermediate (USA), Brent (North Sea), Dubai/Oman (Middle East), and Tapis (Asia- Pacific), which are likely to be highly correlated. This paper analyses the volatility spillover and asymmetric effects across and within the four markets, using three multivariate GARCH models, namely the constant conditional correlation (CCC), vector ARMA-GARCH (VARMA-GARCH) and vector ARMA-asymmetric GARCH (VARMA-AGARCH) models. A rolling window approach is used to forecast the 1-day ahead conditional correlations. The paper presents evidence of volatility spillovers and asymmetric effects on the conditional variances for most pairs of series. In addition, the forecast conditional correlations between pairs of crude oil returns have both positive and negative trends. Moreover, the optimal hedge ratios and optimal portfolio weights of crude oil across different assets and market portfolios are evaluated in order to provide important policy implications for risk management in crude oil markets.
    Keywords: Volatility spillovers, multivariate GARCH, conditional correlation, crude oil prices, spot returns, forward returns, futures returns
    JEL: C22 C32 G32
    Date: 2010–08
  6. By: Rasmus Tangsgaard Varneskov (School of Economics and Management, Aarhus University and CREATES); Valeri Voev (School of Economics and Management, Aarhus University and CREATES)
    Abstract: Recently, consistent measures of the ex-post covariation of financial assets based on noisy high-frequency data have been proposed. A related strand of literature focuses on dynamic models and covariance forecasting for high-frequency data based covariance measures. The aim of this paper is to investigate whether more sophisticated estimation approaches lead to more precise covariance forecasts, both in a statistical precision sense and in terms of economic value. A further issue we address, is the relative importance of the quality of the realized measure as an input in a given forecasting model vs. the model’s dynamic specification. The main finding is that the largest gains result from switching from daily to high-frequency data. Further gains are achieved if a simple sparsesampling covariance measure is replaced with a more efficient and noise-robust estimator.
    Keywords: Forecast evaluation, Volatility forecasting, Portfolio optimization, Mean-variance analysis.
    JEL: C32 C53 G11
    Date: 2010–08–26
  7. By: Rasmus Tangsgaard Varneskov (School of Economics and Management, Aarhus University and CREATES)
    Abstract: This paper considers the performance of di erent long-memory dynamic models when forecasting volatility in the stock market using implied volatility as an exogenous variable in the information set. Observed volatility is sep- arated into its continuous and jump components in a framework that allows for consistent estimation in the presence of market microstructure noise. A comparison between a class of HAR- and ARFIMA models is facilitated on the basis of out-of-sample forecasting performance. Implied volatility conveys incremental information about future volatility in both specifications, improv- ing performance both in- and out-of-sample for all models. Furthermore, the ARFIMA class of models dominates the HAR specications in terms of out-of- sample performance both with and without implied volatility in the information set. A vectorized ARFIMA (vecARFIMA) model is introduced to control for possible endogeneity issues. This model is compared to a vecHAR specication, re-enforcing the results from the single equation framework.
    Keywords: ARFIMA, HAR, Implied Volatility, Jumps, Market Microstructure Noise, VecARFIMA, Volatility Forecasting
    JEL: C14 C22 C32 C53 G10
    Date: 2010–08–19
  8. By: Jeroen V.K. Rombouts (Institute of Applied Economics at HEC Montréal, CIRANO, CIRPEE, Université catholique de Louvain (CORE)); Lars Stentoft (Department of Finance at HEC Montréal, CIRANO, CIRPEE and CREATES)
    Abstract: This paper uses asymmetric heteroskedastic normal mixture models to fit return data and to price options. The models can be estimated straightforwardly by maximum likelihood, have high statistical fit when used on S&P 500 index return data, and allow for substantial negative skewness and time varying higher order moments of the risk neutral distribution. When forecasting out-of-sample a large set of index options between 1996 and 2009, substantial improvements are found compared to several benchmark models in terms of dollar losses and the ability to explain the smirk in implied volatilities. Overall, the dollar root mean squared error of the best performing benchmark component model is 39% larger than for the mixture model. When considering the recent financial crisis this difference increases to 69%.
    Keywords: Asymmetric heteroskedastic models, finite mixture models, option pricing, out-of-sample prediction, statistical fit
    JEL: C11 C15 C22 G13
    Date: 2010–08–24
  9. By: Janine Aron; John Muellbauer
    Abstract: This paper presents new models for aggregate UK data on mortgage possessions (foreclosures) and mortgage arrears (payment delinquencies). The innovations include the treatment of difficuly to observe variations in loan quality and shifts in forbearance policy by lenders, by common latent variables estimated in a system of equations for arrears and possessions, for quarterly data over 1983-2009. A second innovation is the theory-justified use of an estimate of the proportion of mortgages in negative equity, based on an average debt to equity ratio, as one of the key drivers of possessions and arrears. A third is the systematic treatment of measurement bias in the months in arrears measures. Finally, the paper does not impose a proportional long-run relationship between possessions and arrears assumed in the previous UK literature. A range of economic forecast scenarios for forecasts to 2013 reveals the sensitivity of mortgage possessions and arrears to different economic conditions, highlighting potential risks faced by the UK and its mortgage lenders. A comprehensive review of data on arrears and possessions completes the paper.
    Keywords: Foreclosures, Mortgage possessions, Mortgage payment delinquencies, Mortgage arrears, UK mortgage market, Defaults, Unobserved components model
    JEL: G21 G28 R21 C51 C53 E27
    Date: 2010
  10. By: Leonidas Tsiaras (Department of Business Studies, ASB, Aarhus University and CREATES)
    Abstract: Models for the conditional joint distribution of the U.S. Dollar/Japanese Yen and Euro/Japanese Yen exchange rates, from November 2001 until June 2007, are evaluated and compared. The conditional dependency is allowed to vary across time, as a function of either historical returns or a combination of past return data and option-implied dependence estimates. Using prices of currency options that are available in the public domain, risk-neutral dependency expectations are extracted through a copula repre- sentation of the bivariate risk-neutral density. For this purpose, we employ either the one-parameter \Normal" or a two-parameter \Gumbel Mixture" specification. The latter provides forward-looking information regarding the overall degree of covariation, as well as, the level and direction of asymmetric dependence. Specifications that include option-based measures in their information set are found to outperform, in-sample and out-of-sample, models that rely solely on historical returns.
    Keywords: Exchange Rates, Implied Correlation, Copula, Forecasting, Options
    JEL: F31 F37 G14 G15
    Date: 2010–01–12
  11. By: Octavio Fernández-Amador; Josef Baumgartner; Jesús Crespo-Cuaresma
    Abstract: We assess empirically the vertical price transmission mechanism between producer and consumer prices of milk products in Austria using monthly data for the period from January 1996 to February 2010. We consider explicitly the existence of asymmetries in the adjustment to the long-run equilibrium using two different types of threshold vector error correction (VEC) models, where an inaction band in the adjustment to the long-run relationship is defined and alternatively where price dynamics differ between periods of increasing and decreasing trends in causal prices. Our results indicate that asymmetries play an important role in the pass-through of prices for milk products in Austria. We provide statistical evidence concerning the fact that the adjustment only tends to take place when deviations from the equilibrium are large enough. Milk, dairy and cheese products and butter tend to remain in positive margins (measured as deviations from the long-run equilibrium) for the retailers' side. The explicit modeling of nonlinearities does not improve out-of-sample forecasting performance.
    Keywords: Asymmetric price transmission, threshold models, cointegration, milk prices.
    JEL: C32 L11 Q13
    Date: 2010–07
  12. By: Thomas Q. Pedersen (School of Economics and Management, Aarhus University and CREATES)
    Abstract: This paper provides detailed insights into predictability of the entire stock and bond return distribution through the use of quantile regression. This allows us to examine speci?c parts of the return distribution such as the tails or the center, and for a suf?ciently ?ne grid of quantiles we can trace out the entire distribution. A univariate quantile regression model is used to examine stock and bond return distributions individually, while a multivariate model is used to capture their joint distribution. An empirical analysis on US data shows that certain parts of the return distributions are predictable as a function of economic state variables. The results are, however, very different for stocks and bonds. The state variables primarily predict only location shifts in the stock return distribution, while they also predict changes in higher-order moments in the bond return distribution. Out-of-sample analyses show that the relative accuracy of the state variables in predicting future returns varies across the distribution. A portfolio study shows that an investor with power utility can obtain economic gains by applying the empirical return distribution in portfolio decisions instead of imposing an assumption of lognormally distributed returns.
    Keywords: Return predictability, return distribution, quantile regression, multivariate model, out-of-sample forecast, portfolio choice
    JEL: C21 C31 G11 G12
    Date: 2010–07–01

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.