nep-for New Economics Papers
on Forecasting
Issue of 2015‒04‒02
twelve papers chosen by
Rob J Hyndman
Monash University

  1. Modeling and forecasting crude oil price volatility: Evidence from historical and recent data By Lux, Thomas; Segnon, Mawuli; Gupta, Rangan
  2. Predicting the direction of US stock markets using industry returns By Pönkä, Harri
  3. Short- and mid-term forecasting of baseload electricity prices in the UK: The impact of intra-day price relationships and market fundamentals By Katarzyna Maciejowska; Rafal Weron
  4. Exploiting the Errors: A Simple Approach for Improved Volatility Forecasting By Tim Bollerslev; Andrew J. Patton; Rogier Quaedvlieg
  5. Development of Prediction Model of Basic Budget Parameters in Russian Federation By P. A. Nazarov; Kazakova, Maria
  6. Forecasting GDP with global components. This time is different By Hilde C. Bjørnland; Francesco Ravzzolo; Leif Anders Thorsrud
  7. Systematic Errors in Growth Expectations over the Business Cycle By Jonas Dovern; Nils Jannsen
  8. Monitoring the world business cycle By Maximo Camacho; Jaime Martinez-Martin
  9. Comparing the Forecasting Ability of Financial Conditions Indices: The Case of South Africa By Mehmet Balcilar; Rangan Gupta; Renee van Eyden; Kirsten Thompson; Anandamayee Majumdar
  10. Market Inefficiencies and Forecastability of Spot Rates in the Shipping Sector By Nils Wittmann; Eppinger Marcus
  11. Does Joint Modelling of the World Economy Pay Off? Evaluating Global Forecasts from a Bayesian GVAR By Dovern, Jonas; Feldkircher, Martin; Huber , Florian
  12. Oil Price Forecastability and Economic Uncertainty By Stelios Bekiros; Rangan Gupta; Alessia Paccagnini

  1. By: Lux, Thomas; Segnon, Mawuli; Gupta, Rangan
    Abstract: This paper uses the Markov-switching multifractal (MSM) model and generalized autoregressive conditional heteroscedasticity (GARCH)-type models to forecast oil price volatility over the time periods from January 02, 1875 to December 31, 1895 and from January 03, 1977 to March 24, 2014. Based on six different loss functions and by means of the superior predictive ability (SPA) test, we evaluate and compare their forecasting performance at short and long horizons. The empirical results indicate that none of our volatility models can uniformly outperform other models across all six different loss functions. However, the new MSM model comes out as the model that most often across forecasting horizons and subsamples cannot be outperformed by other models, with long memory GARCH-type models coming out second best.
    Keywords: Crude oil prices,GARCH,Multifractal processes,SPA test
    JEL: C52 C53 C22
    Date: 2015
  2. By: Pönkä, Harri
    Abstract: In this paper, we examine the directional predictability of excess stock market returns by lagged excess returns from industry portfolios and a number of other commonly used variables by means of dynamic probit models. We focus on the directional component of the market returns because, for investment purposes, forecasting the direction of return correctly is presumably more relevant than the accuracy of point forecasts. Our findings suggest that only a small number of industries have predictive power for market returns. We also find that the binary response models outperform conventional predictive regressions in forecasting the direction of the market return. Finally, we test trading strategies and find that a number of industry portfolios contain information that can be used to improve investment returns.
    Keywords: industry excess return, sign prediction, probit model, forecasting
    JEL: C25 C53 C58 G17
    Date: 2014–02–24
  3. By: Katarzyna Maciejowska; Rafal Weron
    Abstract: In this paper we investigate whether considering the fine structure of half-hourly electricity prices, the market closing prices of fundamentals (natural gas, coal and CO2) and the system-wide demand can lead to significantly more accurate short- and mid-term forecasts of APX UK baseload prices. We evaluate the predictive accuracy of a number of univariate and multivariate time series models over a three-year out-of-sample forecasting period and compare it against that of a benchmark autoregressive model. We find that in the short-term, up to a few business days ahead, a disaggregated model which independently predicts the intra-day prices and then takes their average to yield baseload price forecasts is the best performer. However, in the mid-term, factor models which explore the correlation structure of intra-day prices lead to significantly (as measured by the Diebold-Mariano test) better baseload price forecasts. At the same time, we observe that the inclusion of fundamental variables - especially natural gas prices (in the short-term) and coal prices (in the mid-term) - provides significant gains. The CO2 prices, on the other hand, generally do not improve the price forecasts at all, at least in the time period considered in this study (Apr. 2009 - Dec. 2013).
    Keywords: Electricity price; Forecasting; Vector autoregression; Factor model; Principal components;
    JEL: C32 C38 C53 Q47
    Date: 2015
  4. By: Tim Bollerslev (Duke University, NBER and CREATES); Andrew J. Patton (Duke University); Rogier Quaedvlieg (Maastricht University)
    Abstract: We propose a new family of easy-to-implement realized volatility based forecasting models. The models exploit the asymptotic theory for high-frequency realized volatility estimation to improve the accuracy of the forecasts. By allowing the parameters of the models to vary explicitly with the (estimated) degree of measurement error, the models exhibit stronger persistence, and in turn generate more responsive forecasts, when the measurement error is relatively low. Implementing the new class of models for the S&P500 equity index and the individual constituents of the Dow Jones Industrial Average, we document significant improvements in the accuracy of the resulting forecasts compared to the forecasts from some of the most popular existing models that implicitly ignore the temporal variation in the magnitude of the realized volatility measurement errors.
    Keywords: Realized volatility, Forecasting, Measurement Errors, HAR, HARQ
    JEL: C22 C51 C53 C58
    Date: 2015–03–10
  5. By: P. A. Nazarov (Russian Presidential Academy of National Economy and Public Administration (RANEPA)); Kazakova, Maria (Gaidar Institute for Economic Policy; Russian Presidential Academy of National Economy and Public Administration)
    Abstract: In this paper, we will construct a simulation and forecast of the main parameters of the Russian federal budget. In particular, we will describe the general scheme of prediction and analysis of statistical data used in the calculations. Next series will be tested for stationarity and cointegration and studied for their statistical characteristics. On the results of this test will be carried out forecast budget guidelines in Russia and ways to improve the quality of the obtained forecasts will be discussed. The calculations will be formulated conclusions.
    Keywords: budget, Russia, modelling, forecast, prediction, data, statistics
    Date: 2014–06
  6. By: Hilde C. Bjørnland (Norges Bank (Central Bank of Norway) and BI Norwegian Business School); Francesco Ravzzolo (Norges Bank (Central Bank of Norway)); Leif Anders Thorsrud (BI Norwegian Business School)
    Abstract: A long strand of literature has shown that the world has become more global. Yet, the recent Great Global Recession turned out to be hard to predict, with forecasters across the world committing large forecast errors. We examine whether knowledge of in-sample co-movement across countries could have been used in a more systematic way to improve forecast accuracy at the national level. In particular, we ask if a model with common international business cycle factors forecasts better than the purely domestic alternative? To answer this question we employ a Dynamic Factor Model (DFM) and run an out-of-sample forecasting experiment. Our results show that exploiting the informational content in a common global business cycle factor improves forecasting accuracy in terms of both point and density forecast evaluation across a large panel of countries. In line with much reported in-sample evidence, we also document that the Great Recession has a huge impact on this result. The event causes a clear preference shift towards the model including a common global factor. Similar shifts are not observed earlier in the evaluation sample. However, this time is different also in other respects. On longer forecasting horizons the performance of the DFM deteriorates substantially in the aftermath of the Great Recession. This indicates that the recession shock itself was felt globally, but that the recovery phase has been very different across countries.
    Keywords: Bayesian Dynamic Factor Model (BDFM), Forecasting, Model uncertainty and global factors
    JEL: C11 C53 F17
    Date: 2015–03–24
  7. By: Jonas Dovern; Nils Jannsen
    Abstract: Using real-time data, we analyze how the systematic expectation errors of professional forecasters in 19 advanced economies depend on the state of the business cycle. Our results indicate that the general result that forecasters systematically overestimate output growth (across all countries) masks considerable differences across different business-cycle states. We show that forecasts for recessions are subject to a large negative systematic forecast error (forecasters overestimate growth), while forecasts for recoveries are subject to a positive systematic forecast error. Forecasts made for expansions have, if anything, a small systematic forecast error for large forecast horizons. When we link information about the business-cycle state in the target year with quarterly information about its state in the forecasting period, we find that forecasters realize business-cycle turning points somewhat late. Using cross-country evidence, we demonstrate that the positive relationship between a change in trend growth rates and forecast bias, as suggested in the literature, breaks down when only focusing on forecasts made for expansions
    Keywords: Macroeconomic expectations, forecasting, forecast bias, survey data
    JEL: C5 E2 E3
    Date: 2015–02
  8. By: Maximo Camacho (Universidad de Murcia); Jaime Martinez-Martin (Banco de España)
    Abstract: We propose a Markov-switching dynamic factor model to construct an index of global business cycle conditions, for performing short-term forecasts of quarterly world GDP growth in real time and computing real-time business cycle probabilities. To overcome the real-time forecasting challenges, the model takes into account mixed frequencies, asynchronous data publication and leading indicators. Our pseudo real-time results show that this approach provides reliable and timely inferences of quarterly world growth and of the state of the world business cycle on a monthly basis.
    Keywords: real-time forecasting, world economic indicators, business cycles, non-linear dynamic factor models
    JEL: E32 C22 E27
    Date: 2015–03
  9. By: Mehmet Balcilar (Department of Economics, Eastern Mediterranean University, Famagusta, North Cyprus, via Mersin 10, Turkey); Rangan Gupta (Department of Economics, University of Pretoria); Renee van Eyden (Department of Economics, University of Pretoria); Kirsten Thompson (Department of Economics, University of Pretoria); Anandamayee Majumdar (Center for Advanced Statistics and Econometrics, Soochow University, Suzhou, China)
    Abstract: In this paper we test the forecasting ability of three estimated financial conditions indices (FCIs) with respect to key macroeconomic variables of output growth, inflation and interest rates. We do this by forecasting the aforementioned macroeconomic variables based on the information contained in the three alternative FCIs using a Bayesian VAR (BVAR), nonlinear logistic vector smooth transition autoregression (VSTAR) and nonparametric (NP) and semi-parametric (SP) regressions, and compare the results with the standard benchmarks of random-walk, univariate autoregressive and classical VAR models. The three FCIs are constructed using rolling-window principal component analysis (PCA), dynamic model averaging (DMA) in the context of a time-varying parameter factor-augmented vector autoregressive (TVP-FAVAR) model, and a time-varying parameter vector autoregressive (TVP-VAR) model with constant factor loadings. Our results suggest that the VSTAR model performs best in the case of forecasting manufacturing production and inflation, while a SP specification proves to be the best for forecasting the interest rate. More importantly, statistics testing for significant differences in forecast errors across models corroborate the finding of superior predictive ability of the nonlinear models.
    Keywords: Financial conditions index, dynamic model averaging, nonlinear logistic smooth transition vector autoregressive model
    JEL: C32 G01 E44 E32
    Date: 2015–03
  10. By: Nils Wittmann (TU Dortmund University); Eppinger Marcus (University of Hamburg)
    Abstract: The market for shipping goods across oceans is particular when it comes to spot shipping rates and its derivatives compared to other asset classes. First, since the underlying of shipping rates is a service, not an asset or commodity, rates cannot be short sold, which leads to an inefficient market. Also, active traders are mainly charteres and ship owners interested in smooth cash flows, rather than speculators, so trading can be thin. Second, cargo shipping rates exhibit a high degree of autocorrelation in its time series. In this paper we empirically take advantage of both characteristics and demonstrate the forecasting ability of Time Charter rates and Futures for future Spot rates in the cargo shipping sector and compare different models. In a sample from 2004 to 2007 with daily data and explicitly the Financial Crisis, we estimate and compare explanatory power and forecast errors of ARIMA processes, Vector Autoregressions and Vector Error Correction Models to simple Random Walks. VAR models outperform all other models in terms of Root Mean Squared Error (RMSE) and can therefore be used to explain and also forecast future Spot rates. Extending an ARIMA model to include Time Charter rates however, does not help to explain future spot rates. Based on these results our models can be used to create trading schemes, which would have outperformed benchmark indices. This result holds, after controlling for transaction costs.
    Keywords: Market Efficiency, Forecasting, Vector Auto Regression, Trading Schemes
    JEL: C53 G14 G17
    Date: 2014–12
  11. By: Dovern, Jonas; Feldkircher, Martin; Huber , Florian
    Abstract: We analyze how modeling international dependencies improves forecasts for the global economy based on a Bayesian GVAR with SSVS prior and stochastic volatility. To analyze the source of performance gains, we decompose the predictive joint density into its marginals and a copula term capturing the dependence structure across countries. The GVAR outperforms forecasts based on country-specific models. This performance is solely driven by superior predictions for the dependence structure across countries, whereas the GVAR does not yield better predictive marginal densities. The relative performance gains of the GVAR model are particularly pronounced during volatile periods and for emerging economies.
    Keywords: GVAR; global economy; forecast evaluation; log score; copula
    Date: 2015–03–27
  12. By: Stelios Bekiros (IPAG Business School, 184 Boulevard Saint-Germain, 75006 Paris, France); Rangan Gupta (Department of Economics, University of Pretoria); Alessia Paccagnini (Department of Economics, Università degli Studi di Milano - Bicocca - Milan)
    Abstract: Information on economic policy uncertainty (EPU) does matter in predicting oil returns especially when accounting for omitted nonlinearities in the relationship between these two variables via a time-varying coefficient approach. In this work, we compare the forecastability of standard, Bayesian and TVP-VAR models against the random-walk and benchmark AR models. Our results indicate that over the period 1900:1-2014:2 the time-varying VAR model with stochastic volatility outranks all alternative models.
    Keywords: Oil prices, economic policy uncertainty, forecasting
    JEL: C22 C32 C53 E60 Q41
    Date: 2015–03

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.