nep-for New Economics Papers
on Forecasting
Issue of 2015‒09‒11
seven papers chosen by
Rob J Hyndman
Monash University

  1. Steady-state priors and Bayesian variable selection in VAR forecasting By Dimitrios P. Louzis
  2. Forecasting Aggregate Retail Sales: The Case of South Africa By Goodness C. Aye; Mehmet Balcilar Author-Name-First Mehmet; Rangan Gupta; Anandamayee Majumdar
  3. On the Forecasting of Financial Volatility Using Ultra-High Frequency Data By António A. F. Santos
  4. Forecasting Electricity Spot Prices using Lasso: On Capturing the Autoregressive Intra-day Structure By Florian Ziel
  5. Republic of Poland: Selected Issues By International Monetary Fund. European Dept.
  6. Systemic Risk, Aggregate Demand, and Commodity Prices By Javier Gómezâ€Pineda; Dominique M. Guillaume; Kadir Tanyeri
  7. Risk Premiums in the Cross-Section of Commodity Convenience Yields By Thomas Bollinger; Axel Kind

  1. By: Dimitrios P. Louzis (Bank of Greece)
    Abstract: This study proposes methods for estimating Bayesian vector autoregressions (VARs) with an automatic variable selection and an informative prior on the unconditional mean or steady-state of the system. We show that extant Gibbs sampling methods for Bayesian variable selection can be efficiently extended to incorporate prior beliefs on the steady-state of the economy. Empirical analysis, based on three major US macroeconomic time series, indicates that the out-of-sample forecasting accuracy of a VAR model is considerably improved when it combines both variable selection and steady-state prior information.
    Keywords: Bayesian VAR, Steady states, Variable selection, Macroeconomic forecasting
    JEL: C32
    Date: 2015–07
  2. By: Goodness C. Aye (Department of Economics, University of Pretoria Author-Email:; Mehmet Balcilar Author-Name-First Mehmet (Department of Economics, Eastern Mediterranean University, Famagusta, Northern Cyprus); Rangan Gupta (Department of Economics, University of Pretoria); Anandamayee Majumdar (Soochow University Center for Advance Statistics and Econometric Research, Suzhou, China)
    Abstract: Forecasting aggregate retail sales may improve portfolio investors’ ability to predict movements in the stock prices of the retailing chains. Therefore, this paper uses 26 (23 single and 3 combination) forecasting models to forecast South Africa’s aggregate seasonal retail sales. We use data from 1970:01–2012:05, with 1987:01-2012:05 as the out-of-sample period. We deviate from the uniform symmetric quadratic loss function typically used in forecast evaluation exercises. Hence, we consider loss functions that overweight forecast error in booms and recessions to check whether a specific model that appears to be a good choice on average is also preferable in times of economic stress. To this end, we use the weighted RMSE and weighted version of the Diebold-Mariano tests to evaluate the different forecasts. Focussing on the single models alone, results show that their performances differ greatly across forecast horizons and for different weighting schemes. However, the combination forecasts models in general produced better forecasts and are largely unaffected by business cycles and time horizons.
    Keywords: seasonality; weighted loss; retail sales forecasting; combination forecasts; South Africa
    JEL: C32 C53 E32
    Date: 2014
  3. By: António A. F. Santos (Faculty of Economics, University of Coimbra, and GEMF, Portugal)
    Abstract: The measurement of the volatility is key in financial markets. This is true not only because decisions are made in an environment of uncertainty, but because sometimes the volatility element overpowers all the remaining aspects in the decision process. Huge movements in the prices of the assets (volatility) can lead to huge losses and also huge gains. There are models to establish the fair prices for certain kind of assets, in that the only parameter that is not directly observable is the parameter characterizing the volatility. However, it is well established in the literature that the evolution of the volatility can be forecasted. Several parametric models have been proposed for modeling the volatility evolution, for example, the Autoregressive Conditional Heteroscedastic (ARCH) and the Stochastic Volatility model (SV). Nowadays, we live in a “Big Data” world, and even for non-professionals of financial markets, it is possible to record data obtained at every second. Recently, measures of volatility have been developed using intraday data, for example, the measure of realized volatility. One of the main aspects to consider is that intraday data and measures of realized volatility are associated with unequal time-spaced observations. In this paper, we compare the forecasts of the volatility evolution using intradaily observations and daily observations, and by trying to conciliate both kind of forecasts, for the data obtained from US and European stock markets, we find out that the use of measures of realized volatility represent an important improvement in volatility forecasting, that can be added to the more well established models that are used in this context, ARCH and SV models.
    Keywords: ARCH models, Big data, Intraday data, Realized volatility, Stochastic volatility.
    JEL: C11 C15 C53 G17
    Date: 2015–08
  4. By: Florian Ziel
    Abstract: In this paper we present a regression based model for day-ahead electricity spot prices. We estimate the considered linear regression model by the lasso estimation method. The lasso approach allows for many possible parameters in the model, but also shrinks and sparsifies the parameters automatically to avoid overfitting. Thus, it is able to capture changes in the intraday dependency structure of the electricity price as the estimated model structure can vary over the day. We discuss in detail the estimation results which provide insights to the intraday behavior of electricity prices. We perform an out-of-sample forecasting study for several European electricity markets. The results illustrate well that the efficient lasso based estimation technique can join advantages from two common model approaches.
    Date: 2015–09
  5. By: International Monetary Fund. European Dept.
    Abstract: This Selected Issues paper employs a suite of models to determine the main drivers of inflation in Poland. Inflation in Poland has stayed below the lower bound of the target band for about two years with external shocks adding to downward pressure during 2014. The paper provides a range of inflation forecasts to assess the likelihood of protracted low inflation. The paper considers the main factors underlying recent inflation developments and assesses the importance of first-round indirect and second-round effects of external shocks for headline inflation. Using a variety of models, the paper also provides possible forecast paths for inflation in Poland.
    Keywords: Poland;inflation, price, prices, price inflation, expectations
    Date: 2015–07–14
  6. By: Javier Gómezâ€Pineda; Dominique M. Guillaume; Kadir Tanyeri
    Abstract: The paper presents a global model with systemic and country risks, as well as commodity prices.We show that systemic risk shocks have an important impact on world economic activity, with the busts in world output gap corresponding to unobserved systemic risk associated with major financial events. In addition, systemic risk shocks are shown to be important drivers of output gaps while country risk premium shocks can have important effects on the trade balance. Commodity prices, in particular the price of oil, are shown to be demand driven. The model performs well at one- and four-quarter horizons compared to a survey of analysts' forecasts. In addition, systemic risk shocks explain a large share of the forecast variance for the world output gap, country output gaps, the price of oil, and country risk premiums. The importance of systemic risk shocks lends support for financial surveillance with a systemic focus.
    Keywords: Central banks and their policies;Foreign exchange;International finance;Systemic risk;Capital flows;Financial linkages, Global imbalances Commodity prices, output, country risk, trade, Open Economy Macroeconomics, Forecasting and Simulation,
    Date: 2015–07–20
  7. By: Thomas Bollinger (Department of Business and Economics, University of Basel, Switzerland); Axel Kind (Department of Economics, University of Konstanz, Germany)
    Abstract: This paper investigates risk premiums embedded in commodity convenience yields, i.e., returns on convenience-claim investments. The analysis is conducted in two steps. First, monthly convenience yields are extracted from a broad sample of commodity futures by using a three-factor model. Second, a multi-factor asset pricing model with conditional betas is estimated to determine risk premiums embedded in convenience-claim returns. The empirical analysis is carried out on monthly cross-sections of 22 commodities in the period from January 1991 to December 2011. It reveals the existence of significant premiums embedded in convenience yields for systematic risk factors typically related to other asset classes. While the predictability of the risk premiums via instrumental variables is limited, changes in conditional betas are found to forecast variations in convenience yields.
    Keywords: Commodity Futures, Convenience Yield, Term Structure, Risk Premiums
    JEL: G12 G13 E44
    Date: 2015–08–10

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.