nep-for New Economics Papers
on Forecasting
Issue of 2016‒01‒18
twelve papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting hierarchical and grouped time series through trace minimization By Shanika L Wickramasuriya; George Athanasopoulos; Rob J Hyndman
  2. Forecasting tourist arrivals to Turkey By Yılmaz, Engin
  3. DSGE model-based forecasting of modelled and nonmodelled inflation variables in South Africa By Rangan Gupta; Patrick T. Kanda; Mampho P. Modise; Alessia Paccagnini
  4. Forecasting Value-at-Risk under Temporal and Portfolio Aggregation By Erik Kole; Thijs Markwat; Anne Opschoor; Dick van Dijk
  5. Predicting Belgium’s GDP using targeted bridge models By Christophe Piette
  6. Predicting US bank failures with internet search volume data By Florian Schaffner
  7. The Evasive Predictive Ability of Core Inflation By Pincheira, Pablo; Selaive, Jorge; Nolazco, Jose Luis
  8. Modelling and forecasting rig rates on the Norwegian Continental Shelf By Terje Skjerpen; Halvor Briseid Storrøsten; Knut Einar Rosendahl; Petter Osmundsen
  9. One-Day Prediction of State of Turbulence for Portfolio. Models for Binary Dependent Variable By Marcin Chlebus
  10. Decomposition of Time Series Data of Stock Markets and its Implications for Prediction: An Application for the Indian Auto Sector By Jaydip Sen; Tamal Datta Chaudhuri
  11. Oil price forecastability and economic uncertainty By Stelios D. Bekiros; Rangan Gupta; Alessia Paccagnini
  12. Long-run evolution of the global economy - Part 2: Hindcasts of innovation and growth By Timothy J. Garrett

  1. By: Shanika L Wickramasuriya; George Athanasopoulos; Rob J Hyndman
    Abstract: Large collections of time series often have aggregation constraints due to product or geographical hierarchies. The forecasts for the disaggregated series are usually required to add up exactly to the forecasts of the aggregated series, a constraint known as “aggregate consistencyâ€. The combination forecasts proposed by Hyndman et al. (2011) are based on a Generalized Least Squares (GLS) estimator and require an estimate of the covariance matrix of the reconciliation errors (i.e., the errors that arise due to aggregate inconsistency). We show that this is impossible to estimate in practice due to identifiability conditions.
    Keywords: Hierarchical time series, forecasting, reconciliation, contemporaneous error correlation, trace minimization
    JEL: C32 C53
    Date: 2015
  2. By: Yılmaz, Engin
    Abstract: Modeling and forecasting techniques of the tourist arrivals are many and diverse. Th ere is no unique model that exactly outperforms the other models in every situation. Actually a few studies have realized modeling and forecasting the tourist arrivals to Turkey and these studies have not focused on the total tourist arrivals. Th ese studies have focused on the tourist arrivals to Turkey country by country (or OECD countries). In addition to this, structural time series models have not been used in modeling and forecasting the tourist arrivals to Turkey. In this sense, this paper is the fi rst study which uses the seasonal autoregressive integrated moving average model and the structural time series model in order to forecast the total tourist arrivals to Turkey. Two diff erent models are developed to forecast the total tourist arrivals to Turkey using monthly data for the period 2002-2013. Th e results of the study show that two models provide accurate predictions but the seasonal autoregressive integrated moving average model produces more accurate short-term forecasts than the structural time series model. It is noted that the seasonal autoregressive integrated moving average model shows a very successful performance in the forecasting the total tourist arrivals to Turkey.
    Keywords: structural time series models; arima; tourist arrivals; tourist demand; Turkey
    JEL: C53
    Date: 2015–12–28
  3. By: Rangan Gupta; Patrick T. Kanda; Mampho P. Modise; Alessia Paccagnini
    Abstract: Inflation forecasts are a key ingredient for monetary policy-making – especially in an inflation targeting country such as South Africa. Generally, a typical Dynamic Stochastic General Equilibrium (DSGE) only includes a core set of variables. As such, other variables, for example alternative measures of inflation that might be of interest to policy-makers, do not feature in the model. Given this, we implement a closed-economy New Keynesian DSGE model-based procedure which includes variables that do not explicitly appear in the model. We estimate such a model using an in-sample covering 1971Q2 to 1999Q4 and generate recursive forecasts over 2000Q1 to 2011Q4. The hybrid DSGE performs extremely well in forecasting inflation variables (both core and nonmodelled) in comparison with forecasts reported by other models such as AR(1). In addition, based on ex-ante forecasts over the period 2012Q1–2013Q4, we find that the DSGE model performs better than the AR(1) counterpart in forecasting actual GDP deflator inflation.
    Keywords: DSGE model; Inflation; Core variables; Noncore variables
    JEL: C11 C32 C53 E27 E47
    Date: 2015
  4. By: Erik Kole (Erasmus University Rotterdam, the Netherlands); Thijs Markwat (Robeco Asset Management, the Netherlands); Anne Opschoor (VU University Amsterdam, the Netherlands); Dick van Dijk (Erasmus University Rotterdam, the Netherlands)
    Abstract: We examine the impact of temporal and portfolio aggregation on the quality of Value-at-Risk (VaR) forecasts over a horizon of ten trading days for a well-diversified portfolio of stocks, bonds and alternative investments. The VaR forecasts are constructed based on daily, weekly or biweekly returns of all constituent assets separately, gathered into portfolios based on asset class, or into a single portfolio. We compare the impact of aggregation to that of choosing a model for the conditional volatilities and correlations, the distribution for the innovations and the method of forecast construction. We find that the degree of temporal aggregation is most important. Daily returns form the best basis for VaR forecasts. Modelling the portfolio at the asset or asset class level works better than complete portfolio aggregation, but differences are smaller. The differences from the model, distribution and forecast choices are also smaller compared to temporal aggregation.
    Keywords: forecast evaluation; aggregation; Value-at-Risk; model comparison
    JEL: C22 C32 C52 C53 G17
    Date: 2015–01–04
  5. By: Christophe Piette (Research Department, NBB)
    Abstract: This paper investigates the usefulness, within the frameworks of the standard bridge model and the ‘bridging with factors’ approach, of a predictor selection procedure that builds on the elastic net algorithm. A pseudo-real time forecasting exercise is performed, in which estimates for Belgium’s quarterly GDP are generated using a monthly dataset of 93 potential predictors. While the simulation results indicate that specifying forecasting models using this procedure can lead to a slight improvement in terms of predictive accuracy over shorter horizons, the forecasting errors made by these ‘targeted’ models are not found to be significantly different from those based on the principal components extracted from the entire set of available indicators. In other words, the only advantage of following such an approach lies in the fact that it enables the forecaster to streamline the information set.
    Keywords: bridge models, nowcasting, variable selection
    JEL: C22 E37
    Date: 2016–01
  6. By: Florian Schaffner
    Abstract: This study investigates how well weekly Google search volumes track and predict bank failures in the United States between 2007 and 2012, contributing to the expanding literature that exploits internet data for the prediction of events. Different duration models with time-varying covariates are estimated. Higher Google search volumes go hand in hand with higher failure rates, and the coefficients for the Google volume growth index are highly significant. However, Google’s predictive power quickly dissipates for future failure rates.
    Keywords: Bank failures, internet, financial crisis, Google, survival analysis
    JEL: G17 G18 G19 G21 G28
    Date: 2015–12
  7. By: Pincheira, Pablo; Selaive, Jorge; Nolazco, Jose Luis
    Abstract: We explore the ability of traditional core inflation –consumer prices excluding food and energy– to predict headline CPI annual inflation. We analyze a sample of OECD and non-OECD economies using monthly data from January 1994 to March 2015. Our results indicate that sizable predictability emerges for a small subset of countries. For the rest of our economies predictability is either subtle or undetectable. These results hold true even when implementing an out-of-sample test of Granger causality especially designed to compare forecasts from nested models. Our findings partially challenge the common wisdom about the ability of core inflation to forecast headline inflation, and suggest a careful weighting of the traditional exclusion of food and energy prices when assessing the size of the monetary stimulus.
    Keywords: Inflation, Forecasting, Time Series, Monetary Policy, Core Inflation
    JEL: E3 E31 E37 E4 E43 E44 E5 E52
    Date: 2016–01–06
  8. By: Terje Skjerpen; Halvor Briseid Storrøsten; Knut Einar Rosendahl; Petter Osmundsen (Statistics Norway)
    Abstract: Knowledge about rig markets is crucial for understanding the global oil market. In this paper we first develop a simple bargaining model for rig markets. Then we examine empirically the most important drivers for rig rate formation of floaters operating at the Norwegian Continental Shelf in the period 1991q4 to 2013q4. We use reduced form time series models with two equations and report conditional point and bootstrapped interval forecasts for rig rates and capacity utilization. We then consider two alternative simulations to examine how the oil price and remaining petroleum reserves influence rig rate formation of floaters. In the first alternative simulation we assume a relatively high crude oil price equal to 100 USD (2010) per barrel for the entire forecast period, whereas the reference case features the actual oil price with extrapolated values for the last quarters in the forecast period. According to our results, the rig rates will be about 34 percent higher in 2016q4 with the higher oil price. In the second alternative simulation we explore the effects of opening the Barents Sea and areas around Jan Mayen for petroleum activity. This contributes to dampening the fall in the rig rates and capacity utilization over the last part of the forecast period.
    Keywords: Rig rates; Capacity utilization; Oil price; Forecasting; Bootstrapping
    JEL: C32 C51 C53 L71 Q47
    Date: 2015–12
  9. By: Marcin Chlebus (Faculty of Economic Sciences, University of Warsaw)
    Abstract: This paper proposes an approach to predict states (states of tranquillity and turbulence) for a current portfolio in a one-day horizon. The prediction is made using 3 different models for a binary variable (LOGIT, PROBIT, CLOGLOG), 4 definitions of a dependent variable (1%, 5%, 10%, 20% of worst realization of returns), 3 sets of independent variables (untransformed data, PCA analysis and factor analysis). Additionally an optimal cut-off point analysis is performed. The evaluation of the models was based on the LR test, Hosmer-Lemeshow test, GINI coefficient analysis and KROC criterion based on the ROC curve. Six combinations of assumptions have been chosen as appropriate (any model for a binary variable, the dependent variable defined as 5% or 10% of worst realization of returns, untransformed data, 5% or 10% cut-off point respectively). Models built on these assumptions meet all the formal requirements and have a high predictive and discriminant ability.
    Keywords: prediction, state of turbulence, regime switching, risk management, risk measure, market risk
    JEL: C53 C58 G17
    Date: 2016
  10. By: Jaydip Sen; Tamal Datta Chaudhuri
    Abstract: With the rapid development and evolution of sophisticated algorithms for statistical analysis of time series data, the research community has started spending considerable effort in technical analysis of such data. Forecasting is also an area which has witnessed a paradigm shift in its approach. In this work, we have used the time series of the index values of the Auto sector in India during January 2010 to December 2015 for a deeper understanding of the behavior of its three constituent components, e.g., the Trend, the Seasonal component, and the Random component. Based on this structural analysis, we have also designed three approaches for forecasting and also computed their accuracy in prediction using suitably chosen training and test data sets. The results clearly demonstrate the accuracy of our decomposition results and efficiency of our forecasting techniques, even in presence of a dominant Random component in the time series.
    Date: 2016–01
  11. By: Stelios D. Bekiros; Rangan Gupta; Alessia Paccagnini
    Abstract: Information on economic policy uncertainty does matter in predicting the change in oil prices. We compare the forecastability of standard, Bayesian and time-varying VAR against univariate models. The time-varying VAR model outranks all alternative models over the period 2007:1–2014:2.
    Keywords: Oil prices; Economic policy uncertainty; Forecasting
    JEL: C22 C32 C53 E60 Q41
    Date: 2015–07
  12. By: Timothy J. Garrett
    Abstract: Long-range climate forecasts use integrated assessment models to link the global economy to greenhouse gas emissions. This paper evaluates an alternative economic framework outlined in part 1 of this study (Garrett, 2014) that approaches the global economy using purely physical principles rather than explicitly resolved societal dynamics. If this model is initialized with economic data from the 1950s, it yields hindcasts for how fast global economic production and energy consumption grew between 2000 and 2010 with skill scores > 90 % relative to a model of persistence in trends. The model appears to attain high skill partly because there was a strong impulse of discovery of fossil fuel energy reserves in the mid-twentieth century that helped civilization to grow rapidly as a deterministic physical response. Forecasting the coming century may prove more of a challenge because the effect of the energy impulse appears to have nearly run its course. Nonetheless, an understanding of the external forces that drive civilization may help development of constrained futures for the coupled evolution of civilization and climate during the Anthropocene.
    Date: 2016–01

This nep-for issue is ©2016 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.