nep-for New Economics Papers
on Forecasting
Issue of 2015‒10‒25
thirteen papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting with Temporal Hierarchies By George Athanasopoulos; Rob J Hyndman; Nikolaos Kourentzes; Fotios Petropoulos
  2. Forecasting Inflation: Phillips Curve Effects on Services Price Measures By Tallman, Ellis W.; Zaman, Saeed
  3. Macroeconomic Factors and Equity Premium Predictability By Buncic, Daniel; Tischhauser, Martin
  4. Consensus forecasters: How good are they individually and why? By Franses, Ph.H.B.F.; Maassen, N.
  5. Real-Time Forecasting with a Large, Mixed Frequency, Bayesian VAR By McCracken, Michael W.; Owyang, Michael T.; Sekhposyan, Tatevik
  6. Economic Implications of Enhanced Forecast Accuracy: The Case of Photovoltaic Feed-In Forecasts By Ruhnau, Oliver; Hennig, Patrick; Madlener, Reinhard
  7. Inflation as a Global Phenomenon—Some Implications for Policy Analysis and Forecasting By Ayse Kabukcuoglu; Enrique Martínez-García
  8. Financial and Real Sector Leading Indicators of Recessions in Brazil using Probabilistic Models By Fernando N. de Oliveira
  9. The Predictability of cay and cayMS for Stock and Housing Returns: A Nonparametric Causality in Quantile Test By Mehmet Balcilar; Rangan Gupta; Ricardo M. Sousa; Mark E. Wohar
  10. Predicting Stock Returns and Volatility with Investor Sentiment Indices: A Reconsideration using a Nonparametric Causality-in-Quantiles Test By Mehmet Balcilar; Rangan Gupta; Clement Kyei
  11. “Should I stay or should I go?”: Weather forecasts and the economics of “short breaks” By L. Zirulia
  12. The evolution of inflation expectations in Canada and the US By James Yetman
  13. Forecasting the volatility of the Japanese stock market using after-hour information in other markets By Nirodha I Jayawardenaa; Neda Todorova; Bin Li; Jen-Je Su

  1. By: George Athanasopoulos; Rob J Hyndman; Nikolaos Kourentzes; Fotios Petropoulos
    Abstract: This paper introduces the concept of Temporal Hierarchies for time series forecasting. A temporal hierarchy can be constructed for any time series by means of non-overlapping temporal aggregation. Predictions constructed at all aggregation levels are combined with the proposed framework to result in temporally reconciled, accurate and robust forecasts. The implied combination mitigates modelling uncertainty, while the reconciled nature of the forecasts results in a unified prediction that supports aligned decisions at different planning horizons: from shortterm operational up to long-term strategic planning. The proposed methodology is independent of forecasting models. It can embed high level managerial forecasts that incorporate complex and unstructured information with lower level statistical forecasts. Our results show that forecasting with temporal hierarchies increases accuracy over conventional forecasting, particularly under increased modelling uncertainty. We discuss organisational implications of the temporally reconciled forecasts using a case study of Accident & Emergency departments.
    Keywords: Hierarchical forecasting, temporal aggregation, reconciliation, forecast combination
    JEL: C44 C53
    Date: 2015
  2. By: Tallman, Ellis W. (Federal Reserve Bank of Cleveland); Zaman, Saeed (Federal Reserve Bank of Cleveland)
    Abstract: We estimate an empirical model of inflation that exploits a Phillips curve relationship between a measure of unemployment and a subaggregate measure of inflation (services). We generate an aggregate inflation forecast from forecasts of the goods subcomponent separate from the services subcomponent, and compare the aggregated forecast to the leading time-series univariate and standard Phillips curve forecasting models. Our results indicate notable improvements in forecasting accuracy statistics for models that exploit relationships between services inflation and the unemployment rate. In addition, models of services inflation using the short-term unemployment rate (less than 27 weeks) as the real economic indicator display additional modest forecast accuracy improvements.
    Keywords: Inflation forecasting; Phillips curve; disaggregated inflation forecasting models; trend-cycle model
    JEL: C22 C53 E31 E37
    Date: 2015–10–14
  3. By: Buncic, Daniel; Tischhauser, Martin
    Abstract: Neely et al. (2014) have recently demonstrated how to efficiently combine information from a set of popular technical indicators together with the standard Goyal andWelch (2008) predictor variables widely used in the equity premium forecasting literature to improve outof- sample forecasts of the equity premium using a small number of principal components. We show that forecasts of the equity premium can be further improved by, first, incorporating broader macroeconomic data into the information set, second, improving the selection of the most relevant factors and combining the most relevant factors by means of a forecast combination regression, and third, imposing theoretically motivated positivity constraints on the forecasts of the equity premium. Applying standard out-of-sample forecast evaluation tests, we find that in particular our proposed forecast combination approach, which combines forecasts of the most relevant Neely et al. (2014) and macroeconomic factors and further imposes positivity constraints on the equity premium forecasts, generates statistically significant and economically sizable improvements over the best performing model of Neely et al. (2014). Out-of-sample R2 values can be as high as 1.75%, with (annualised) gains in certainty equivalent returns of up to 3.35%, relative to the ALL factors forecasts of Neely et al. (2014).
    Keywords: Equity premium predictability, Factor models, Macroeconomic variables, Adaptive Lasso, Sign restrictions, Forecast combination, Asset allocation
    JEL: G12 G17 C53 E44
    Date: 2015–10
  4. By: Franses, Ph.H.B.F.; Maassen, N.
    Abstract: We analyze the monthly forecasts for annual US GDP growth, CPI inflation rate and the unemployment rate delivered by forty professional forecasters collected in the Consensus database for 2000M01-2014M12. To understand why some forecasters are better than others, we create simple benchmark model-based forecasts. Evaluating the individual forecasts against the model forecasts is informative for how the professional forecasters behave. Next, we link this behavior to forecast performance. We find that forecasters who impose proper judgment to model-based forecasts also have highest forecast accuracy, and hence, they do not perform best just by luck.
    Keywords: macroeconomic forecasts, expert adjustment
    JEL: E27 E37
    Date: 2015–10–12
  5. By: McCracken, Michael W. (Federal Reserve Bank of St. Louis); Owyang, Michael T. (Federal Reserve Bank of St. Louis); Sekhposyan, Tatevik (Texas A&M University)
    Abstract: We assess point and density forecasts from a mixed-frequency vector autoregression (VAR) to obtain intra-quarter forecasts of output growth as new information becomes available. The econometric model is specified at the lowest sampling frequency; high frequency observations are treated as different economic series occurring at the low frequency. We impose restrictions on the VAR to account explicitly for the temporal ordering of the data releases. Because this type of data stacking results in a high-dimensional system, we rely on Bayesian shrinkage to mitigate parameter proliferation. The relative performance of the model is compared to forecasts from various time-series models and the Survey of Professional Forecaster's. We further illustrate the possible usefulness of our proposed VAR for causal analysis.
    Keywords: Vector autoregression; Blocking model; Stacked vector autoregression; Mixed-frequency estimation; Bayesian methods; Nowcasting; Forecasting
    JEL: C22 C52 C53
    Date: 2015–10–08
  6. By: Ruhnau, Oliver (RWTH Aachen University); Hennig, Patrick (Grundgrün Energie GmbH); Madlener, Reinhard (E.ON Energy Research Center, Future Energy Consumer Needs and Behavior (FCN))
    Abstract: Forecasts are usually evaluated in terms of accuracy. With regard to application, the question arises if the most accurate forecast is also optimal in terms of forecast related costs and risks. Combining insights from research and practice, we show that this is indeed not necessarily the case. Our analysis is grounded in the dynamic field of short-term forecasting of solar electricity feed-in. A clear sky model is implemented and combined with a linear model, an autoregressive model, and an artificial neural network. These models are applied to a portfolio of ten large-scale photovoltaic systems in Germany. We compare the different models in order to quantify the connection between errors and costs. We find that apart from accuracy, correlation with market prices is an important characteristic of forecasts when economic implications are considered as important.
    Keywords: Forecasting evaluation; renewable energy; electricity markets; balancing costs; artificial neural networks; clear sky model; Germany
    Date: 2015–06
  7. By: Ayse Kabukcuoglu (Koc University); Enrique Martínez-García (Federal Reserve Bank of Dallas & Southern Methodist University)
    Abstract: We evaluate the performance of inflation forecasts based on the open-economy Phillips curve by exploiting the spatial pattern of international propagation of inflation. We model these spatial linkages using global inflation and either domestic slack or oil price fluctuations, motivated by a novel interpretation of the forecasting implications of the workhorse open-economy New Keynesian model (Martínez-García and Wynne (2010), Kabukcuoglu and Martínez-García (2014)). We find that incorporating spatial interactions yields significantly more accurate forecasts of local inflation in 14 advanced countries (including the U.S.) than a simple autoregressive model that captures only the temporal dimension of the inflation dynamics.
    Keywords: Inflation Dynamics; Open-Economy Phillips Curve; Forecasting.
    JEL: C21 C23 C53 F41
    Date: 2015–10
  8. By: Fernando N. de Oliveira
    Abstract: We examine the usefulness of various financial and real sector variables to forecast recessions in Brazil between one and eight quarters ahead. We estimate probabilistic models of recession and select models based on their out-of-sample forecasts, using the Receiver Operating Characteristic (ROC) function. We find that the predictive out-of-sample ability of several models vary depending on the numbers of quarters ahead to forecast and on the number of regressors used in the model specification. The models selected seem to be relevant to give early warnings of recessions in Brazil
    Date: 2015–09
  9. By: Mehmet Balcilar (Eastern Mediterranean University, Turkey and University of Pretoria, South Africa); Rangan Gupta (Department of Economics, University of Pretoria); Ricardo M. Sousa (Department of Economics, University of Minho, Campus of Gualtar, 4710-057 - Braga - Portugal); Mark E. Wohar (University of Nebraska-Omaha, USA and Loughborough University, UK)
    Abstract: We use a nonparametric causality-in-quantiles test to compare the predictive ability of cay and cayMS for excess and real stock and housing returns and their volatility using quarterly data for the US over the periods of 1952:Q1-2014:Q3 and 1953:Q2-2014:Q3 respectively. Our results reveal strong evidence of nonlinearity and regime changes in the relationship between asset returns and cay or cayMS, which corroborates the relevance of this econometric framework. Moreover, we confirm the outperformance of cayMS vis-à-vis cay and their relevance for excess stock returns. Furthermore, we show that cayMS is particularly useful at forecasting certain quantiles of the conditional distribution. As for housing returns, the empirical evidence suggests that the predictive ability of cay and cayMS is relatively low. Yet, cay outperforms cayMS over the majority of the quantiles of the conditional distribution of the variance of real housing returns.
    Keywords: stock returns, housing returns, quantile, nonparametric, causality
    JEL: C32 C53 Q41
    Date: 2015–10
  10. By: Mehmet Balcilar (Department of Economics, Eastern Mediterranean University, Famagusta, via Mersin 10, Northern Cyprus, Turkey and Department of Economics, University of Pretoria); Rangan Gupta (Department of Economics, University of Pretoria); Clement Kyei (Department of Economics, University of Pretoria)
    Abstract: Evidence of monthly stock returns predictability based on popular investor sentiment indices, namely SBW and SPLS as introduced by Baker and Wurgler (2006, 2007) and Huang et al. (2015) respectively are mixed. While, linear predictive models show that only SPLS can predict excess stock returns, nonparametric models (which accounts for misspecification of the linear frameworks due to nonlinearity and regime changes) finds no evidence of predictability based on either of these two indices for not only stock returns, but also its volatility. However, in this paper, we show that when we use a more general nonparametric causality-in –quantiles model of Balcilar et al., (2015), in fact, both SBW and SPLS can predict stock returns and its volatility, with SPLS being a relatively stronger predictor of excess returns during bear and bull regimes, and SBW being a relatively powerful predictor of volatility of excess stock returns, barring the median of the conditional distribution.
    Keywords: Investor sentiment, stock markets, linear causality, nonlinear dependence, nonparametric causality, causality-in-quantiles
    JEL: C22 C32 C53 G02 G10 G11 G17
    Date: 2015–10
  11. By: L. Zirulia
    Abstract: The aim of this paper is to model the decisions of tourists and a monopolist firm when weather forecasts are available. Before deciding whether to go on holiday or not, but after the firm has decided and posted its price, tourists can look at weather forecasts. Our results show that the price chosen by the firm and the corresponding equilibrium profit are decreasing as a function of the accuracy of weather forecasts. Consumers, instead, are better off the more accurate weather forecasts become. Managerial and policy implications are also derived.
    JEL: D83 L12 L83
    Date: 2015–10
  12. By: James Yetman
    Abstract: We model inflation forecasts as monotonically diverging from an estimated long-run anchor point towards actual inflation as the forecast horizon shortens. Fitting the model with forecaster-level data for Canada and the US, we identify three key differences between the two countries. First, the average estimated anchor of US inflation forecasts has tended to decline gradually over time in rolling samples, from 3.4% for 1989-1998 to 2.2% for 2004-2013. By contrast, it has remained close to 2% since the mid-1990 for Canadian forecasts. Second, the variance of estimates of the long-run anchor is considerably lower for the panel of Canadian forecasters than US ones following Canada's adoption of inflation targets. And third, forecasters in Canada look much more alike than those in the US in terms of the weight that they place on the anchor. One explanation for these results is that an explicit inflation targeting regime (Canada) provides for less uncertainty about future monetary policy actions than a monetary policy regime where there was no explicit numerical inflation target (the US before 2012) to anchor expectations.
    Keywords: Inflation expectations, decay function, inflation targeting
    Date: 2015–10
  13. By: Nirodha I Jayawardenaa; Neda Todorova; Bin Li; Jen-Je Su
    Keywords: High Frequency, Realized Volatility, Overnight volatility, Forecasting
    Date: 2015–08

This nep-for issue is ©2015 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.