nep-for New Economics Papers
on Forecasting
Issue of 2012‒02‒08
six papers chosen by
Rob J Hyndman
Monash University

  1. Evaluating point and density forecasts of DSGE models By Wolters, Maik Hendrik
  2. Forecasting Data Vintages By Tara M. Sinclair
  3. The Impact of Seasonal and Price Adjustments on the Predictability of German GDP Revisions By Jens Boysen-Hogrefe, Stefan Neuwirth
  4. Agricultural Banking and Early Warning Models for the Bank Failures of the Late 2000s Great Recession By Li, Xiaofei; Escalante, Cesar L.; Epperson, James E.; Gunter, Lewell F.
  5. Predicting soil erosion risk at the Alqueva dam watershed By Ferreira, Vera; Panagopoulos, Thomas
  6. Nonlinear expectations in speculative markets - Evidence from the ECB survey of professional forecasters By Reitz, Stefan; Rülke, Jan-Christoph; Stadtmann, Georg

  1. By: Wolters, Maik Hendrik
    Abstract: This paper investigates the accuracy of point and density forecasts of four DSGE models for inflation, output growth and the federal funds rate. Model parameters are estimated and forecasts are derived successively from historical U.S. data vintages synchronized with the Fed’s Greenbook projections. Point forecasts of some models are of similar accuracy as the forecasts of nonstructural large dataset methods. Despite their common underlying New Keynesian modeling philosophy, forecasts of different DSGE models turn out to be quite distinct. Weighted forecasts are more precise than forecasts from individual models. The accuracy of a simple average of DSGE model forecasts is comparable to Greenbook projections for medium term horizons. Comparing density forecasts of DSGE models with the actual distribution of observations shows that the models overestimate uncertainty around point forecasts.
    Keywords: DSGE models; forecasting; model uncertainty; forecast combination; density forecasts; real-time data; Greenbook
    JEL: E0 E32 C53 E31 E37
    Date: 2012–01–23
  2. By: Tara M. Sinclair (George Washington University)
    Abstract: This article provides a discussion of Clements and Galvão’s “Forecasting with Vector Autoregressive Models of Data Vintages: US output growth and inflation.” Clements and Galvão argue that a multiple-vintage VAR model can be useful for forecasting data that are subject to revisions. Clements and Galvão draw a “distinction between forecasting future observations and revisions to past data,” which brings yet another real time data issue to the attention of forecasters. This comment discusses the importance of taking data revisions into consideration and compares the multiple-vintage VAR approach of Clements and Galvão to a state-space approach.
    Keywords: Real time data, Evaluating forecasts, Forecasting practice, Time series, Econometric models
    JEL: C53
    Date: 2012–01
  3. By: Jens Boysen-Hogrefe, Stefan Neuwirth
    Abstract: Releases of the GDP are subject to revisions over time. This paper examines the predictability of German GDP revisions using forecast rationality tests. Previous studies of German GDP covering data until 1997 finds that revisions of real seasonally adjusted GDP are predictable. This paper uses a newly available real-time data to analyze the revisions of real seasonal adjusted GDP, of nominal unadjusted GDP, of the seasonal pattern, and of the GDP deflator for the period between 1992 and 2006. We find that the revisions of the nominal unadjusted GDP are unpredictable, but that the revisions of the price adjustments are predictable. Nevertheless, revisions of real seasonally adjusted GDP are hardly predictable and less well predictable compared to earlier studies. This lower predictability seems to be linked to the finding that revisions of seasonal adjustments are hardly predictable, too, and that their predictability decreased over time
    Keywords: Real-time data, GDP revisions, noise, news, forecasting, seasonal adjustment, price adjustment
    JEL: C82
    Date: 2012–01
  4. By: Li, Xiaofei; Escalante, Cesar L.; Epperson, James E.; Gunter, Lewell F.
    Abstract: This paper is designed to validate if the agricultural sector can once again be labeled as an instigator of the late-2000s Great Recession using the early warning models technique. The empirical results indicate that exposure to agribusiness operations does not necessarily enhance a banksâ tendency to fail.
    Keywords: Agricultural Banking, Early warning signals, In-sample accuracy, Out-of-sample forecasting, Agricultural Finance, Research Methods/ Statistical Methods, G21, G32, G33, C01,
    Date: 2012
  5. By: Ferreira, Vera (cieo - research centre for spatial and organizational dynamics); Panagopoulos, Thomas (cieo - research centre for spatial and organizational dynamics)
    Abstract: Soil erosion is serious economic and environmental concern. Assessing soil erosion risk in the Alqueva dam watershed is urgently needed to conserve soil and water resources and prevent the accelerated dam siltation, taking into account the possible land-use changes, due to tourism development, intensification of irrigated farming and biomass production, as well as climate change. A comprehensive methodology that integrates Revised Universal Soil Loss Equation (RUSLE) model and Geographic Information Systems (GIS) with geostatistical techniques was adopted to study different land-use and management scenarios. The main objective of this study stage is to determine the soil erosion vulnerability of an agro-silvo pastoral system. The resultant soil erosion map shows an average of 14.1 t/ha/ year, with serious erosion risk (higher than 50 t/ha/year) in 4.3% of area. The highest values are associated mainly to high slopes and low vegetation. The final prediction maps for soil erosion and for each factor considered, can be used as a solid base to create a Decision Support System so as to provide specific procedures for decision-makers, promoting for sustainability of the ecosystems, reducing the risk of erosion and consequently increase lifetime of dam, under various land use and management scenarios
    Keywords: Soil Erosion; Land-use; Geostatistic; RUSLE; Geographic Information System
    JEL: Q01 Q15 Q24
    Date: 2012–01–23
  6. By: Reitz, Stefan (Institute for Quantitative Business and Economics Research); Rülke, Jan-Christoph (Department of Economics); Stadtmann, Georg (Department of Business and Economics)
    Abstract: Chartist and fundamentalist models have proven to be capable of replicating stylized facts on speculative markets. In general, this is achieved by specifying nonlinear interactions of otherwise linear asset price expectations of the respective trader groups. This paper investigates whether or not regressive and extrapolative expectations themselves exhibit significant nonlinear dynamics. The empirical results are based on a new data set from the European Central Bank Survey of Professional Forecasters on oil price expectations. In particular, we find that forecasters form destabilizing expectations in the neighborhood of the fundamental value, whereas expectations tend to be stabilizing in the presence of substantial oil price misalignment.
    Keywords: Agent based models; nonlinear expectations; survey data
    JEL: C33 D84 F31
    Date: 2012–01–03

This nep-for issue is ©2012 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.