nep-for New Economics Papers
on Forecasting
Issue of 2010‒12‒18
ten papers chosen by
Rob J Hyndman
Monash University

  1. Term Structure Models Can Predict Interest Rate Volatility. But How? By Hideyuki Takamizawa
  2. Evaluating Combined Non-Replicable Forecasts By Chia-Lin Chang; Philip Hans Franses; Michael McAleer
  3. Combining the forecasts in the ECB survey of professional forecasters: can anything beat the simple average? By Véronique Genre; Geoff Kenny; Aidan Meyler; Allan Timmermann
  4. House Market in Chinese Cities: Dynamic Modeling, In-Sampling Fitting and Out-of-Sample Forecasting By Leung, Charles Ka Yui; Chow, Kenneth; Yiu, Matthew; Tam, Dickson
  5. Real-time Inflation Forecast Densities from Ensemble Phillips Curves By Anthony Garratt; James Mitchell; Shaun P. Vahey; Elizabeth C. Wakerly
  6. A Conditionally Heteroskedastic Global Inflation Model By Leonardo Morales-Arias; Guilherme V. Moura
  7. Nowcasting By Marta Bańbura; Domenico Giannone; Lucrezia Reichlin
  8. Construction industry forecasting system dynamic model By Skribans, Valerijs
  9. Investments model development with the system dynamic method By Skribans, Valerijs
  10. Using Experts for Predicting Continuous Outcomes By J. Kivinen; M. Warmuth

  1. By: Hideyuki Takamizawa
    Abstract: This paper attempts to predict the volatility of interest rates through dynamic term structure models. For this attempt, the models are improved, based on the three-factor Gaussian model, to have level-dependent volatilities supported by data. The empirical results show that the predictive power of the proposed models is higher than that of the affine models. Compared with time-series models, it is low for the four-week forecasting horizon but can be comparable for middle to long term rates by extending the horizon up to 32 weeks. The combination of these two different types of forecasts can lead to higher predictive power.
    Date: 2010–11
  2. By: Chia-Lin Chang (Department of Applied Economics, Department of Finance, National Chung Hsing University); Philip Hans Franses (Econometric Institute, Erasmus School of Economics, Erasmus University Rotterdam); Michael McAleer (Erasmus University Rotterdam, Tinbergen Institute, The Netherlands, and Institute of Economic Research, Kyoto University)
    Abstract: Macroeconomic 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.
    JEL: C53 C22 E27 E37
    Date: 2010–12
  3. By: Véronique Genre (European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Geoff Kenny (European Central Bank, DG Research, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Aidan Meyler (European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Allan Timmermann (Rady School of Management and Department of Economics, University of California, San Diego, USA.)
    Abstract: In this paper, we explore the potential gains from alternative combinations of the surveyed forecasts in the ECB Survey of Professional Forecasters. Our analysis encompasses a variety of methods including statistical combinations based on principal components analysis and trimmed means, performance-based weighting, least squares estimates of optimal weights as well as Bayesian shrinkage. We provide a pseudo real–time out-of-sample performance evaluation of these alternative combinations and check the sensitivity of the results to possible data-snooping bias. The latter robustness check is also informed using a novel real time meta selection procedure which is not subject to the data-snooping critique. For GDP growth and the unemployment rate, only few of the forecast combination schemes are able to outperform the simple equal-weighted average forecast. Conversely, for the inflation rate there is stronger evidence that more refined combinations can lead to improvement over this benchmark. In particular, for this variable, the relative improvement appears significant even controlling for data snooping bias. JEL Classification: C22, C53.
    Keywords: forecast combination, forecast evaluation, data snooping, real-time data, Survey of Professional Forecasters.
    Date: 2010–12
  4. By: Leung, Charles Ka Yui; Chow, Kenneth; Yiu, Matthew; Tam, Dickson
    Abstract: This paper attempts to contribute in several ways. Theoretically, it proposes simple models of house price dynamics and construction dynamics, all based on forward-looking agents’ maximization problems, which may carry independent interests. Simplified version of the model implications are estimated with the data from four major cities in China. Both price and construction dynamics exhibit strong persistence in al cities. Significant heterogeneity across cities is found. Our models out-perform widely used alternatives in in-sample-fitting for all cities, although similar success only limited to highly developed cities in out-of-sample forecasting. Policy implications and future research directions are also discussed.
    Keywords: pre-sale; production constraint; collateral constraint; cross-city heterogeneity; fundamental versus policy
    JEL: R30 E30 C33
    Date: 2010
  5. By: Anthony Garratt; James Mitchell; Shaun P. Vahey; Elizabeth C. Wakerly
    Abstract: We examine the effectiveness of recursive-weight and equal-weight combination strategies for forecasting using many time-varying models of the relationship be- tween inflation and the output gap. The forecast densities for inflation reflect the uncertainty across models using many statistical measures of the output gap, and allow for time-variation in the ensemble Phillips curves. Using real-time data for the US, Australia, New Zealand and Norway, we find that the recursive-weight strategy performs well, consistently giving well-calibrated forecast densities. The equal-weight strategy generates poorly-calibrated forecast densities for the US and Australian samples. There is little difference between the two strategies for our New Zealand and Norwegian data. We also find that the ensemble modelling approach performs more consistently with real-time data than with revised data in all four countries.
    JEL: C32 C53 E37
    Date: 2010–12
  6. By: Leonardo Morales-Arias; Guilherme V. Moura
    Abstract: This article proposes a multivariate model of inflation with conditionally heteroskedastic common and country-specific components. The model is estimated in one-step via Quasi-Maximum Likelihood for the G7 countries for the period Q1-1960 to Q4-2009. It is found that various model specifications considered fit well the first and second order dynamics of inflation in the G7. The estimated volatility of the common inflation component captures the international effects of the ‘Great Moderation’ and of the ‘Great Recession’. The model also shows promising capabilities for forecasting inflation in several countries
    Keywords: Global inflation, conditional heteroskedasticity, inflation forecasting
    JEL: E31 E37 F41
    Date: 2010–11
  7. By: Marta Bańbura (European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Domenico Giannone (Université libre de Bruxelles, ECARES, Avenue Roosevelt CP 114 Brussels, Belgium and CEPR.); Lucrezia Reichlin (London Business School and CEPR.)
    Abstract: We define nowcasting as the prediction of the present, the very near future and the very recent past. Crucial in this process is to use timely monthly information in order to nowcast key economic variables, such as e.g. GDP, that are typically collected at low frequency and published with long delays. Until recently, nowcasting had received very little attention by the academic literature, although it was routinely conducted in policy institutions either through a judgemental process or on the basis of simple models. We argue that the nowcasting process goes beyond the simple production of an early estimate as it essentially requires the assessment of the impact of new data on the subsequent forecast revisions for the target variable. We design a statistical model which produces a sequence of nowcasts in relation to the real time releases of various economic data. The methodology allows to process a large amount of information, as it is traditionally done by practitioners using judgement, but it does it in a fully automatic way. In particular, it provides an explicit link between the news in consecutive data releases and the resulting forecast revisions. To illustrate our ideas, we study the nowcast of euro area GDP in the fourth quarter of 2008. JEL Classification: E52, C53, C33.
    Keywords: Nowcasting, News, Factor Model, Forecasting.
    Date: 2010–12
  8. By: Skribans, Valerijs
    Abstract: In a paper construction branch forecasting model which allows to estimate the industry development problems is shown. Difference from anthers models, in given paper the main attention is turned to the building of the living area. Model stands from sub model (blocks): amount of apartments, real estate prices, necessity of apartments and living area forecasting models. Their essence and necessity are shown in separate sections. In the paper final, produced model is applied practical, are given model forecasts. The apartments amount forecasting model show supposition, in which, if there is deficit of dwellings in the economic system, then, first of all, are finance and construct apartments with little areas, i.e., multistory buildings with one-room apartments. Apartment’s amount increase depends from the building financing, as also from middle apartment area and building costs of square meter. The real estate prices forecasting model show supposition, that the once certain price is actual while it is not changed with prices influential parameters. The prices influential parameters are: extend (or diminish) of the live fund, the total market influence on separate market segments (and vice versa). The apartment’s necessities forecasting model show supposition, that in beginning a certain volume of necessities remains not changed, while on their has not changed with influential factors. Between the influential factors there are: apartments amount increase – it diminish apartments amount necessities; apartments wear (amount diminish) - it increase apartments amount necessities; improve of live circumstances - it diminish apartments amount necessities. The live area forecasting model is an additional model, which provides functioning of another’s models. Shown supposition was related with a population tries to purchase apartments with such properties which is similar to present at the market apartments. Till with it the middle area in each analyse group is not changes. In not changes middle area circumstances, knowing apartments amount, it is possible to calculate the total live area. Accumulating statistical data it is to adjoin with a problem, that not all data was certain, for a few data only possible borders of oscillations are certain. Till with that, in the paper are examined a few possible situation development variants. Buildings and real estate market development simulated on separate its segments and in an aggregate. Analysing the forecast data for aggregate apartments building in Latvia, in basic scenario it is certain, that the counterbalanced apartments building volume is about 1800 apartments in the year, but noticing oscillations of overproduces and necessities, this size can brief hesitate from 1434 to 2019. A sensitiveness analysis confirms basic scenario, complements it and extends with credible horizon borders.
    Keywords: construction demand; housing fund; system dynamic; market modeling; simulation; economic forecasting
    JEL: C51 C53 C02 L85 L74 C49 C01
    Date: 2010
  9. By: Skribans, Valerijs
    Abstract: In the paper model of macroeconomic turnover and its possibilities for investments modelling are shown. The model consists from four blocks: in the first the theoretical model is described. In the second the model is reflected in accordance with the requirements of system dynamics method, there are shown included influences intercommunications and equalizations. The third block examines demand and supply model for capital and investments. Fourth part complements previous parts, describes additional indexes. The method is offered for using both in forecasting and in teaching.
    Keywords: investment; aggregated demand; macroeconomics; simulation model; system dynamics
    JEL: C51 C52 C82 E22 C70 E21 C53 E27 E30 C50 E23 E00 E37 E20 C01
    Date: 2010
  10. By: J. Kivinen; M. Warmuth
    Date: 2010–12–10

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.