nep-for New Economics Papers
on Forecasting
Issue of 2014‒03‒30
fourteen papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting the Oil-gasoline Price Relationship: Should We Care about the Rockets and the Feathers? By Andrea Bastianin; Marzio Galeotti; Matteo Manera
  2. Artificial Neural Networks and Aggregate Consumption Patterns in New Zealand By Daniel Farhat
  3. Prospects for the use of modern versions of the SNA in the formation of an information base for macroeconomic analysis and forecasting By Alexander Abroskin
  4. Implications of Cost Behavior for Analysts’ Earnings Forecasts By Mustafa Ciftci; Raj Mashruwala; Dan Weiss
  5. Forecasting the US Economy with a Factor-Augmented Vector Autoregressive DSGE model By Stelios Bekiros; Alessia Paccagnini
  6. Conditional Forecasts and Scenario Analysis with Vector Autoregressions for Large Cross-Sections By Martha Banbura; Domenico Giannone; Michèle Lenza
  7. Bootstrap prediction intervals for linear, nonlinear, and nonparametric autoregressions By Pan, Li; Politis, Dimitris N
  8. A note on using the Hodrick-Prescott filter in electricity markets By Rafal Weron; Michal Zator
  9. OECD Forecasts During and After the Financial Crisis: A Post Mortem By Nigel Pain; Christine Lewis; Thai-Thanh Dang; Yosuke Jin; Pete Richardson
  10. The Stock Market, the Real Economy and Contagion By Dirk G Baur; Isaac Miyakawa
  11. Testing for Leverage Effect in Financial Returns. By Christophe Chorro; Dominique Guegan; Florian Ielpo; Hanjarivo Lalaharison
  12. Loan Default Prediction in Ukrainian Retail Banking By Goriunov Dmytro; Venzhyk Katerina
  13. Statistical inference for measures of predictive success By Demuynck T.
  14. Carbon price analysis using empirical mode decomposition By Bangzhu Zhu; Ping Wang; Julien Chevallier; Yiming Wei

  1. By: Andrea Bastianin (University of Milan and FEEM); Marzio Galeotti (University of Milan and IEFE-Bocconi); Matteo Manera (University of Milan-Bicocca and FEEM)
    Abstract: According to the Rockets and Feathers hypothesis (RFH), the transmission mechanism of positive and negative changes in the price of crude oil to the price of gasoline is asymmetric. Although there have been many contributions documenting that downstream prices are more reactive to increases than to decreases in upstream prices, little is known about the forecasting performance of econometric models incorporating asymmetric price transmission from crude oil to gasoline. In this paper we fill this gap by comparing point, sign and probability forecasts from a variety of Asymmetric-ECM (A-ECM) and Threshold Autoregressive ECM (TAR-ECM) specifications against a standard ECM. Forecasts from A-ECM and TAR-ECM subsume the RFH, while the ECM implies symmetric price transmission from crude oil to gasoline. We quantify the forecast accuracy gains due to incorporating the RFH in predictive models for the prices of gasoline and diesel. We show that the RFH is useless for point forecasting, while it can be exploited to produce more accurate sign and probability forecasts. Finally, we highlight that the forecasting performance of the estimated models is time-varying.
    Keywords: Asymmetries, Forecast Evaluation, Gasoline, Crude Oil, Rockets and Feathers
    JEL: C22 C32 C53 Q40 Q47
    Date: 2014–03
  2. By: Daniel Farhat (Department of Economics, University of Otago, New Zealand)
    Abstract: This study engineers a household sector where individuals process macroeconomic information to reproduce consumption spending patterns in New Zealand. To do this, heterogeneous artificial neural networks (ANNs) are trained to forecast changes in consumption. In contrast to existing literature, results suggest that there exists a trained ANN that significantly outperforms a linear econometric model at out-of-sample forecasting. To improve the accuracy of ANNs using only in-sample information, methods for combining private knowledge into social knowledge are explored. For one type of ANN, relying on an expert is beneficial. For most ANN structures, weighting an individual's forecast according to how frequently that individual's ANN is a top performer during in-sample training produces more accurate social forecasts. By focusing only on recent periods, considering the severity of an individual's errors in weighting their forecast is also beneficial. Possible avenues for incorporating ANN structures into artificial social simulation models of consumption are discussed.
    Keywords: Artificial neural networks, forecasting, aggregate consumption, social simulation
    JEL: C45 E17 E27
    Date: 2014–03
  3. By: Alexander Abroskin (Gaidar Institute for Economic Policy)
    Abstract: The article concerns the topical questions of the use of the System of National Accounts (SNA) modern version in the development the information base for macroeconomic analysis and forecasting. The paper examines the main innovations presented in the methodology of the SNA 2008 and supporting further extension of the data base for the development of analytical estimates and short-, medium- and long-term macroeconomic forecasts. The paper investigates in detail the distinctive features of use of the SNA modern version in analyzing the elements of non-financial assets, financial services and international relations in globalized economy. As well are discussed the prospects for the expansion of the analytical indicators system and the development of new aspects of macroeconomic analysis and forecasting based on the methodology of the 2008 SNA.
    Keywords: innovation processes, information support, macroeconomic analysis, methodology, non-observed activity, non-financial assets, forecasting, System of National Accounts, financial assets.
    JEL: O11
    Date: 2014
  4. By: Mustafa Ciftci; Raj Mashruwala; Dan Weiss
    Abstract: Recent work in management accounting offers several novel insights into firms’ cost behavior. This study explores whether financial analysts appropriately incorporate information on two types of cost behavior in predicting earnings - cost variability and cost stickiness. Since analysts’ utilization of information is not directly observable, we model the process of earnings prediction to generate empirically testable hypotheses. The results indicate that analysts “converge to the average” in recognizing both cost variability and cost stickiness, resulting in substantial and systematic earnings forecast errors. Particularly, we find a clear pattern - inappropriate incorporation of available information on cost behavior in earnings forecasts leads to larger errors in unfavorable scenarios than in favorable ones. Overall, enhancing analysts’ awareness of the expense side is likely to improve their earnings forecasts, mainly when sales turn to the worse.
    Keywords: cost stickiness, cost variability, analysts’ earnings forecasts, expense forecasts
    JEL: M41 G12
  5. By: Stelios Bekiros; Alessia Paccagnini
    Abstract: Although policymakers and practitioners are particularly interested in DSGE models, these are typically too stylized to be applied directly to the data and often yield weak prediction re- sults. Very recently, hybrid DSGE models have become popular for dealing with some of the model misspecifications. Ma jor advances in estimation methodology could allow these models to outperform well-known time series models and effectively deal with more complex real-world prob- lems as richer sources of data become available. ln this study we introduce a Bayesian approach to estimate a novel Factor Augmented DSGE model that extends the model of Consolo et al. [Consolo, A., Favero, C.A., and Paccagnini, A., 2009. On the Statistical ldentification of DSGE Models. Journal of Econometrics 150, 99-115]. We perform a comparative predictive evaluation of many different specifications of estimated DSGE models and various classes of VAR models, using datasets from the US economy. Simple and hybrid DSGE models are implemented, such as DSGE-VAR and tested against standard, Bayesian and Factor Augmented VARs. The results can be useful for macro-forecasting and monetary policy analysis.
    Keywords: Forecasting, Marginal data density, DSGE-FAVAR
    JEL: C32 C11 C15
    Date: 2014–02–25
  6. By: Martha Banbura; Domenico Giannone; Michèle Lenza
    Abstract: This paper describes an algorithm to compute the distribution of conditional forecasts,i.e. projections of a set of variables of interest on future paths of some othervariables, in dynamic systems. The algorithm is based on Kalman filtering methods andis computationally viable for large vector autoregressions (VAR) and dynamic factormodels (DFM). For a quarterly data set of 26 euro area macroeconomic and financialindicators, we show that both approaches deliver similar forecasts and scenario assessments.In addition, conditional forecasts shed light on the stability of the dynamicrelationships in the euro area during the recent episodes of financial turmoil and indicatethat only a small number of sources drive the bulk of the fluctuations in the euroarea economy.
    Keywords: vector autoregression; bayesian shrinkage; dynamic factor model; conditional forecast; large cross-sections
    JEL: C11 C13 C33 C53
    Date: 2014–03
  7. By: Pan, Li; Politis, Dimitris N
    Abstract: In order to construct prediction intervals without the combersome--and typically unjustifiable--assumption of Gaussianity, some form of resampling is necessary. The regression set-up has been well-studies in the literature but time series prediction faces additional difficulties. The paper at hand focuses on time series that can be modeled as linear, nonlinear or nonparametric autoregressions, and develops a coherent methodology for the constructuion of bootstrap prediction intervals. Forward and backward bootstrap methods for using predictive and fitted residuals are introduced and compared. We present detailed algorithms for these different models and show that the bootstrap intervals manage to capture both sources of variability, namely the innovation error as well as essimation error. In simulations, we compare the prediction intervals associated with different methods in terms of their acheived coverage level and length of interval. 
    Keywords: Physical Sciences and Mathematics, Confidence intervals, forecasting, time series
    Date: 2014–01–01
  8. By: Rafal Weron; Michal Zator
    Abstract: Recently, Nowotarski et al. (2013) have found that wavelet-based models for the long-term seasonal component (LTSC) are not only better in extracting the LTSC from a series of spot electricity prices but also significantly more accurate in terms of forecasting these prices up to a year ahead than the commonly used monthly dummies and sine-based models. However, a clear disadvantage of the wavelet-based approach is the increased complexity of the technique as compared to the other two classes of LTSC models, which may render it too complicated for practitioners. To facilitate this problem, we propose here a much simpler, yet equally powerful method for identifying the LTSC in electricity spot price series. It makes use of the Hodrick-Prescott (HP) filter, a widely-recognized tool in macroeconomics.
    Keywords: Hodrick-Prescott filter; Electricity spot price; Long-term seasonal component; Robust modeling;
    JEL: C14 C51 C53 Q47
    Date: 2014–03–20
  9. By: Nigel Pain; Christine Lewis; Thai-Thanh Dang; Yosuke Jin; Pete Richardson
    Abstract: This paper assesses the OECD’s projections for GDP growth and inflation during the global financial crisis and recovery, focussing on lessons that can be learned. The projections repeatedly over-estimated growth, failing to anticipate the extent of the slowdown and later the weak pace of the recovery – errors made by many other forecasters. At the same time, inflation was stronger than expected on average. Analysis of the growth errors shows that the OECD projections in the crisis years were larger in countries with more international trade openness and greater presence of foreign banks. In the recovery, there is little evidence that an underestimate of the impact of fiscal consolidation contributed significantly to forecast errors. Instead, the repeated conditioning assumption that the euro area crisis would stabilise or ease played an important role, with growth weaker than projected in European countries where bond spreads were higher than had been assumed. But placing these errors in a historical context illustrates that the errors were not without precedent: similar-sized errors were made in the first oil price shock of the 1970s. In response to the challenges encountered in forecasting in recent years and the lessons learnt, the OECD and other international organisations have sought to improve their forecasting techniques and procedures, to improve their ability to monitor near-term developments and to better account for international linkages and financial market developments. Prévisions de l'OCDE pendant et après la crise financière : Post mortem Ce document évalue les projections de l'OCDE relatives à la croissance du PIB et à l'inflation durant la crise financière mondiale et lors de la reprise, tout en mettant l'accent sur les leçons qui peuvent être tirées. Les projections ont surestimé la croissance de façon répétée, à défaut d'anticiper l'ampleur du ralentissement puis, plus tard, le faible rythme de la reprise — des erreurs commises par de nombreux autres prévisionnistes. Simultanément, l'inflation a été, en moyenne, plus forte que prévu. L'analyse des erreurs relatives à la croissance montre que les prévisions de l'OCDE durant les années de crise économiques ont été plus importantes dans les pays dotés d'une plus grande ouverture au commerce international et d'une plus grande présence de banques étrangères. Durant la reprise, il y a peu d'évidences qu'une sous-estimation de l'impact de la consolidation budgétaire ait conduit de manière significative aux erreurs. Au lieu de cela, l'hypothèse de conditionnement répétée que la crise de la zone euro devrait se stabiliser ou a joué un rôle important, avec une croissance plus faible que prévu dans les pays européens où les écarts de rendement des obligations étaient plus élevés que ce qui avait été supposé. Mais placer ces erreurs dans un contexte historique montre que les erreurs ne sont pas sans précédent: des erreurs de taille similaire ont été faites lors du premier choc des prix du pétrole dans les années 70. En réponse aux difficultés rencontrées dans les prévisions au cours des dernières années et les leçons apprises, l'OCDE et d'autres organisations internationales ont cherché à améliorer leurs techniques et procédures de prévision, afin d'améliorer leur capacité à surveiller l'évolution à court terme et à mieux appréhender les liens internationaux et l'évolution du marché financier
    Keywords: fiscal policy, inflation, forecasting, economic fluctuations, economic outlook, performance économique, prévisions, fluctuations économiques, inflation, politique budgétaire
    JEL: E17 E27 E31 E32 E37 E62 E66 F47 G01
    Date: 2014–03–17
  10. By: Dirk G Baur (Finance Discipline Group, UTS Business School, University of Technology, Sydney); Isaac Miyakawa (Finance Discipline Group, UTS Business School, University of Technology, Sydney)
    Abstract: In this paper we analyze the link between stock market performance and macroe conomic performance for a large number of countries. We study the short-run and long-run relationships and find that stock market returns do not coherently predict future macroeconomic changes for the majority of countries, i.e. the estimates vary considerably both across prediction horizons and across countries. Moreover, we test whether the financial and real economy dynamic linkages increased in the financial crisis in 2008 implying “macro-financial” contagion. The crisis-specific analysis of macro-financial linkages broadens the perspective of existing studies of financial contagion. Our findings indicate that the stock market does not merely reflect future economic conditions but also influences them justifying policy responses as witnessed during the 2008 financial and economic crisis.
    Keywords: global stock markets; real economic activity; predictive regressions; contagion; financial crises; co-integration
    JEL: C22 C32 E44 G01 G14 G15 G18
    Date: 2014–01–01
  11. By: Christophe Chorro (Centre d'Economie de la Sorbonne); Dominique Guegan (Centre d'Economie de la Sorbonne - Paris School of Economics); Florian Ielpo (Lombard Odier Darier Hentsch & Cie - Suisse); Hanjarivo Lalaharison (Centre d'Economie de la Sorbonne)
    Abstract: This article questions the empirical usefulness of leverage effects to describe the dynamics of equity returns. Using a recursive estimation scheme that accurately disentangles the asymmetry coming from the conditional distribution of returns and the asymmetry that is related to the past return to volatility component in GARCH models, we test for the statistical significance of the latter. Relying on both in and out of sample tests we consistently find a weak contribution of leverage effect over the past 25 years of S&P 500 returns, casting light on the importance of the conditional distribution in time series models.
    Keywords: Maximum likelihood method, related-GARCH process, recursive estimation method, mixture of Gaussian distributions, generalized hyperbolic distributions, S&P 500, forecast, leverage effect.
    JEL: C58 C13
    Date: 2014–02
  12. By: Goriunov Dmytro; Venzhyk Katerina
    Abstract: Using a large proprietary dataset provided by the tenth largest Ukrainian banking institution, we posit reasons for loan defaults within two major groups of retail borrowers; car loans and mortgages. Two model types were used, namely logistic regression and neural networks. The results of our estimations suggest that a) data currently collected by banks are sufficient to predict defaults, but bankers should collect more information, and that b) the neural networks model slightly outperforms the logit model in predictive power.
    JEL: G21 G32 G33
    Date: 2013–05–23
  13. By: Demuynck T. (GSBE)
    Abstract: We provide statistical inference for measures of predictive success. These measures are frequently used to evaluate and compare the performance of different models of individual and group decision making in experimental and revealed preference studies. We provide a brief illustration of our findings by comparing the predictive success of different revealed preference tests for models of intertemporal decision making.
    Keywords: Econometric and Statistical Methods and Methodology: General; Design of Experiments: General; Consumer Economics: Empirical Analysis;
    JEL: C10 C90 D12
    Date: 2014
  14. By: Bangzhu Zhu; Ping Wang; Julien Chevallier; Yiming Wei
    Abstract: Mastering the underlying characteristics of carbon price changes can help governments formulate correct policies to keep efficient operation of carbon markets, and investors take effective measures to evade their investment risks. Empirical mode decomposition (EMD), a self-adaption data analysis approach for nonlinear and non-stationary time series, can accurately explain the formation mechanism of carbon price by decomposing it into several intrinsic mode functions (IMFs) and one residue from different scales. In this study, we apply EMD to the European Union Emissions Trading Scheme (EU ETS) carbon price analysis. First, the carbon price is decomposed into eight IMFs and one residue. Moreover, these IMFs and residue are reconstructed into a high frequency component, a low frequency component and a trend component using hierarchical clustering method. The economic meanings of these three components are identified as short term market fluctuations, effects of significant trend breaks, and a long-term trend, respectively. Finally, some strategies are proposed for carbon price forecasting.
    Keywords: Carbon Price; Empirical Mode Decomposition; Multiscale Analysis; Forecasting; EU ETS
    Date: 2014–02–25

This nep-for issue is ©2014 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.