nep-for New Economics Papers
on Forecasting
Issue of 2013‒03‒23
sixteen papers chosen by
Rob J Hyndman
Monash University

  1. Conditiona l Forecast Selection from Many Forecasts: An Application to the Yen/Dollar Exchange Rate By Kei Kawakami
  2. Evaluation of Probabilistic Forecasts: Proper Scoring Rules and Moments By Tsyplakov, Alexander
  3. Evaluating the accuracy of forecasts from vector autoregressions By Todd E. Clark; Michael W. McCracken
  4. Forecasting the USD/EUR daily and monthly rate with machine learning techniques By Papadimitriou, Theophilos; Gogas, Periklis; Plakandaras, Vasilios
  5. It’s not just for inflation: The usefulness of the median CPI in BVAR forecasting By Brent Meyer; Saeed Zaman
  6. Food versus Fuel: Causality and Predictability in Distribution By Andrea Bastianin; Marzio Galeotti; Matteo Manera
  7. Structural-break models under mis-specification: implications for forecasting By Boonsoo Koo; Myung Hwan Seo
  8. Forecasting the insolvency of U.S. banks using Support Vector Machines (SVM) based on Local Learning Feature Selection By Gogas, Periklis; Papadimitriou, Theophilos; Plakandaras, Vasilios
  9. The determinants of macroeconomic forecasts and the Stability and Growth Pact By Patricia Martins; Leonida Correia
  10. Do economies stall? The international evidence By Wai-Yip Alex Ho; James Yetman
  11. Testing for common cycles in non-stationary VARs with varied frecquency data By Hecq A.W.; Urbain J.R.Y.J.; Götz T.B.
  12. Inflation persistence and the rationality of inflation expectations By Sophocles N. Brissimis; Petros M. Migiakis
  13. Revisiting the Empirical Inconsistency of the Permanent Income Hypothesis: Evidence from Rural China By Liping Gao; Hyeongwoo Kim; Yaoqi Zhang
  14. On the causality and determinants of energy and electricity demand in South Africa: A review By Anastassios Pouris
  15. Predictability of stock market activity using Google search queries By Sofía B. Ramos; Helena Veiga; Pedro Latoeiro
  16. House prices, expectations, and time-varying fundamentals By Paolo Gelain; Kevin J. Lansing

  1. By: Kei Kawakami
    Abstract: This paper proposes a new method for forecast selection from a pool of many forecasts. The method uses conditional information as proposed by Giacomini and White (2006). It also extends their pairwise switching method to a situation with many forecasts. I apply the method to the monthly yen/dollar exchange rate and show empirically that my method of switching forecasting models reduces forecast errors compared with a single model.
    Keywords: Conditional predictive ability; Exchange rate; Forecasting; Forecast combinations; Model selection
    JEL: C52 C53 F31 F37
    Date: 2013
  2. By: Tsyplakov, Alexander
    Abstract: The paper provides an overview of probabilistic forecasting and discusses a theoretical framework for evaluation of probabilistic forecasts which is based on proper scoring rules and moments. An artificial example of predicting second-order autoregression and an example of predicting the RTSI stock index are used as illustrations.
    Keywords: probabilistic forecast; forecast calibration; probability integral transform; scoring rule; moment condition
    JEL: C52 C53
    Date: 2013–03–18
  3. By: Todd E. Clark; Michael W. McCracken
    Abstract: This paper surveys recent developments in the evaluation of point and density forecasts in the context of forecasts made by Vector Autoregressions. Specific emphasis is placed on highlighting those parts of the existing literature that are applicable to direct multi-step forecasts and those parts that are applicable to iterated multi-step forecasts. This literature includes advancements in the evaluation of forecasts in population (based on true, unknown model coefficients) and the evaluation of forecasts in the finite sample (based on estimated model coefficients). The paper then examines in Monte Carlo experiments the finite-sample properties of some tests of equal forecast accuracy, focusing on the comparison of VAR forecasts to AR forecasts. These experiments show the tests to behave as should be expected given the theory. For example, using critical values obtained by bootstrap methods, tests of equal accuracy in population have empirical size about equal to nominal size.
    Keywords: Economic forecasting ; Vector autoregression
    Date: 2013
  4. By: Papadimitriou, Theophilos (Democritus University of Thrace, Department of International Economic Relations and Development); Gogas, Periklis (Democritus University of Thrace, Department of International Economic Relations and Development); Plakandaras, Vasilios (Democritus University of Thrace, Department of International Economic Relations and Development)
    Abstract: We combine signal processing to the forecasting abilities of machine learning methods by introducing a hybrid Ensemble Empirical Mode Decomposition (EEMD), Multivariate Adaptive Regression Splines (MARS) and Support Vector Regression (SVR) model in order to forecast monthly and daily US/Euro FX rate. After trend extraction for the initial dataset, MARS selects the most informative variables among the plethora included in our data sample which are then fed into an SVR model. The overall process is repeated twice, one for the trend and one for the fluctuation component of all time series. The above implementation proved its superior forecasting abilities in predicting USD/EUR exchange rate compared to various models both on monthly and daily forecasting horizon. Overall the proposed model a) is a combination of empirically proven effective techniques in forecasting time series, b) is data driven, c) relies on minimum initial assumptions and d) provides a structural aspect of the forecasting problem.
    Keywords: Exchange rate forecasting; Support Vector Regression; local learning; feature selection; Ensemble Empirical Mode Decomposition; time series; trend
    JEL: G15
    Date: 2013–03–19
  5. By: Brent Meyer; Saeed Zaman
    Abstract: In this paper we investigate the forecasting performance of the median CPI in a variety of Bayesian VARs (BVARs) that are often used for monetary policy. Until now, the use of trimmed-mean price statistics in forecasting inflation has often> been relegated to simple univariate or “Philips-Curve” approaches, thus limiting their usefulness in applications that require consistent forecasts of multiple macro variables. We find that inclusion of an extreme trimmed-mean measure—the> median CPI—significantly improves the forecasts of both headline and core CPI. across our wide-ranging set of BVARs. While the inflation forecasting improvements are perhaps not surprising given the current literature on core inflation statistics, we also find that inclusion of the median CPI improves the forecasting> accuracy of the central bank’s primary instrument for monetary policy—the federal funds rate. We conclude with a few illustrative exercises that highlight the usefulness of using the median CPI.>
    Keywords: Bayesian statistical decision theory ; Forecasting ; Monetary policy ; Simulation modeling
    Date: 2013
  6. By: Andrea Bastianin; Marzio Galeotti; Matteo Manera
    Abstract: This paper examines the relationship between biofuels and commodity food prices in the U.S. from a new perspective. While a large body of literature has tried to explain the linkages between sample means and volatilities associated with ethanol and agricultural price returns, little is known about their whole distributions. We focus on predictability in distribution by asking whether ethanol returns can be used to forecast different parts of field crops returns distribution, or vice versa. Density forecasts are constructed using Conditional Autoregressive Expectile models estimated with Asymmetric Least Squares. Forecast evaluation relies on quantile-weighed scoring rules, which identify regions of the distribution of interest to the analyst. Results show that both the centre and the left tail of the ethanol returns distribution can be predicted by using field crops returns. On the contrary, there is no evidence that ethanol can be used to forecast any region of the field crops distribution.
    Keywords: Biofuels, Ethanol, Field Crops, Density Forecasting, Granger Causality, Quantiles
    JEL: C22 C53 Q13 Q42 Q47
    Date: 2013–03
  7. By: Boonsoo Koo; Myung Hwan Seo
    Abstract: This paper shows that in the presence of model mis-specification, the conventional inference procedures for structural-break models are invalid. In doing so, we establish new distribution theory for structural break models under the relaxed assumption that our structural break model is the best linear approximation of the true but unknown data generating process. Our distribution theory involves cube-root asymptotics and it is used to shed light on forecasting practice. We show that the conventional forecasting methods do not necessarily produce the best forecasts in our setting. We also propose a new forecasting strategy, which incorporates our new distribution theory, and apply our forecasting method to numerous macroeconomic data. The performance of various contemporary forecasting methods is compared to ours.
    Keywords: structural breaks, forecasting, mis-specification, cube-root asymptotics, bagging
    Date: 2013
  8. By: Gogas, Periklis (Democritus University of Thrace, Department of International Economic Relations and Development); Papadimitriou, Theophilos (Democritus University of Thrace, Department of International Economic Relations and Development); Plakandaras, Vasilios (Democritus University of Thrace, Department of International Economic Relations and Development)
    Abstract: We propose a Support Vector Machine (SVM) based structural model in order to forecast the collapse of banking institutions in the U.S. using publicly disclosed information from their financial statements on a four-year rolling window. In our approach, the optimum input variable set is defined from a large dataset using an iterative relevance-based selection procedure. We train an SVM model to classify banks as solvent and insolvent. The resulting model exhibits significant ability in bank default forecasting.
    Keywords: Bank insolvency; SVM; local learning; feature selection;
    JEL: G21
    Date: 2013–03–19
  9. By: Patricia Martins; Leonida Correia
    Abstract: This paper identifies the determinants of macroeconomic forecasts (budget balance, public debt and real GDP growth), of the governments of the 15 EU countries. We have used the forecasts of the Stability and Convergence Programmes submitted between 1998/99 and 2008/09 and the European Commission’s. Results show that, in general, economic growth forecasts submitted by European governments are more optimistic than those published by the European Commission. The lack of accuracy of government forecasts is due to “misinformation” regarding the economic situation at the time of their publication. The differences between observed and forecast changes of budget balance and public debt are explained by the output growth forecast errors and the forecasts of the changes in the two fiscal indicators. These forecast changes tend to revise downwards the changes submitted in the previous Program. Therefore, the governments’ “bad intention” seems to result from their lack of commitment to the objectives of previous programs and it explains the recurrent delays in the implementation of their fiscal consolidation plans.
    Keywords: European Union, Stability and Growth Pact, forecast errors
    JEL: E62 H6
    Date: 2013–02
  10. By: Wai-Yip Alex Ho; James Yetman
    Abstract: A "stalling" economy has been defined as one that experiences a discrete deterioration in economic performance following a decline in its growth rate to below some threshold level. Previous efforts to identify stalls have focused primarily on the US economy, with the threshold level being chosen endogenously, and have suggested that the concept of a stall may be useful for macroeconomic forecasting. We examine the international evidence for stalling in a panel of 51 economies using two different definitions of a stall threshold (time-invariant and related to lagged average growth rates) and two complementary empirical approaches (insample statistical significance and out-of-sample forecast performance). We find that the evidence for stalling based on time-invariant thresholds is limited: only 12 of the 51 economies in our sample experience statistically significant stalls, and including a stall threshold generally results in only modest improvements to out-ofsample forecast performance. When we instead model the stall threshold as varying with average growth rates, the number of economies with statistically-significant stalls actually declines (to nine), but in 71% of the cases we examine, including a stall threshold results in an improvement in out-of-sample forecast performance.
    Keywords: business cycles, stall speed, Markov switching
    Date: 2013–03
  11. By: Hecq A.W.; Urbain J.R.Y.J.; Götz T.B. (GSBE)
    Abstract: This paper proposes a new way for detecting the presence of common cyclical features when several time series are observed/sampled at different frequencies, hence generalizing the common-frequency approach introduced by Engle and Kozicki (1993) and Vahid and Engle (1993). We start with the mixed-frequency VAR representation investigated in Ghysels (2012) for stationary time series. For non-stationary time series in levels, we show that one has to account for the presence of two sets of long-run relationships. The First set is implied by identities stemming from the fact that the differences of the high-frequency I(1) regressors are stationary. The second set comes from possible additional long-run relationships between one of the high-frequency series and the low-frequency variables. Our transformed VECM representations extend the results of Ghysels (2012) and are very important for determining the correct set of variables to be used in a subsequent common cycle investigation. This has some empirical implications both for the behavior of the test statistics as well as for forecasting. Empirical analyses with the quarterly real GNP and monthly industrial production indices for, respectively, the U.S. and Germany illustrate our new approach. This is also investigated in a Monte Carlo study, where we compare our proposed mixed-frequency models with models stemming from classical temporal aggregation methods.
    Keywords: Regional and Urban History: General;
    Date: 2013
  12. By: Sophocles N. Brissimis (University of Piraeus); Petros M. Migiakis (Bank of Greece)
    Abstract: The rational expectations hypothesis for survey and model-based inflation forecasts ? from the Survey of Professional Forecasters and the Greenbook respectively ? is examined by properly taking into account persistence in the data. The finding of near-unit-root effects in inflation and inflation expectations motivates the use of a local-to-unity specification of the inflation process that enables us to test whether the data are generated by locally non-stationary or stationary processes. Thus, we test, rather than assume, stationarity of near-unit-root processes. In addition, we set out an empirical framework for assessing relationships between locally non-stationary series. In this context, we test the rational expectations hypothesis by allowing the co-existence of a long-run relationship obtained under the rational expectations restrictions with short-run "learning" effects. Our empirical results indicate that the rational expectations hypothesis holds in the long run, while forecasters adjust their expectations slowly in the short run. This finding lends support to the hypothesis that the persistence of inflation comes from the dynamics of expectations.
    Keywords: Inflation; rational expectations; high persistence
    JEL: C50 E31 E52
    Date: 2013–01
  13. By: Liping Gao; Hyeongwoo Kim; Yaoqi Zhang
    Abstract: Chow (1985) reports strong evidence in favor of the permanent income hypothesis (PIH) using observations from 1953 to 1982 in China. We revisit this issue with rural area household data in China during the post economic reform regime (1978-2009) as well as the postwar US data for comparison. Our in-sample analysis provides strong evidence against the PIH for both countries. Out-of-sample forecast exercises also reveal that consumption changes are highly predictable. Our vector autoregressive (VAR) model analysis also shows significantly positive responses of consumption to income shocks, and non-negligible proportions of variations in consumption are explained by innovations in income.
    Keywords: Permanent Income Hypothesis; Consumption; Generalized Method of Moments; Diebold-Mariano-West Statistic; Vector Autoregressive
    JEL: E21 E27
    Date: 2013–03
  14. By: Anastassios Pouris (Institute for Technological Innovation, University of Pretoria)
    Abstract: The purpose of this paper is to review, summarise and critically assess the academic studies that have dealt with either the causal relationship between energy consumption and growth or the determinants of energy demand in South Africa from 2007 and outline recent forecasts for electricity demand. The results of this review aim to identify gaps in the existing research. From a policy point of view, the findings of this effort have the potential to inform the relevant stakeholders to make appropriate interventions to improve the status quo of the energy sector. The results have indicated that studies examining the causality direction between energy (electricity) consumption and economic growth have failed to reach a consensus. The main differences identified were the time periods examined, the econometric approaches and the variables included in the estimations. Another potential reason for the results is the availability –or lack thereof– of data specific for the country. On the other side, the studies looking at the factor affecting energy (electricity) demand have agreed that economic growth or income or output are considered significant factors. The role of prices was debatable among different studies. This has become more apparent when reviewing the few forecasting efforts in the country that resulted in conflicting results.
    Keywords: Review, South Africa, energy sector, causality, determinants
    Date: 2013–03
  15. By: Sofía B. Ramos; Helena Veiga; Pedro Latoeiro
    Abstract: This paper analyzes whether web search queries predict stock market activity in a sample of the largest European stocks. We provide evidence that i) an increase in web searches for stocks on Google engine is followed by a temporary increase in volatility and volume and a drop in cumulative returns. ii) An increase for web search queries for the market index leads to a decrease in the returns of the index as well as of the stock index futures and an increase in implied volatility. iii) Attention interacts with behavioral biases. The predictability of web searches for return and liquidity is enhanced when firm prices and market prices hit a 52-week high and diminished when the market hits a 52-week low. iv) Investors tend to process more market information than firm specific information in investment decisions, confirming limited attention theory.
    Keywords: Behavioral Finance, Google Search Volume Index, Investor Attention, Predictability
    Date: 2013–03
  16. By: Paolo Gelain; Kevin J. Lansing
    Abstract: We investigate the behavior of the equilibrium price-rent ratio for housing in a standard asset pricing model. We allow for time-varying risk aversion (via external habit formation) and time-varying persistence and volatility in the stochastic process for rent growth, consistent with U.S. data for the period 1960 to 2011. Under fully-rational expectations, the model significantly underpredicts the volatility of the U.S. price-rent ratio for reasonable levels of risk aversion. We demonstrate that the model can approximately match the volatility of the price-rent ratio in the data if near-rational agents continually update their estimates for the mean, persistence and volatility of fundamental rent growth using only recent data (i.e., the past 4 years), or if agents employ a simple moving-average forecast rule that places a large weight on the most recent observation. These two versions of the model can be distinguished by their predictions for the correlation between expected future returns on housing and the price-rent ratio. Only the moving-average model predicts a positive correlation such that agents tend to expect higher future returns when house prices are high relative to fundamentals–a feature that is consistent with survey evidence on the expectations of real-world housing investors.
    Keywords: Asset pricing ; Housing - Prices
    Date: 2013

This nep-for issue is ©2013 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.