nep-for New Economics Papers
on Forecasting
Issue of 2008‒12‒01
eight papers chosen by
Rob J Hyndman
Monash University

  1. Predicting the Signs of Forecast Errors By Nazaria Solferino; Robert J. Waldmann
  2. Performance of Analyst´ Earnings Forecasting - Evidence from the Finnish Emerging Markets 1987-2005 By Mikko Kepsu; Hannu Schadewitz; Markku Vieru
  3. Improving Feeder Cattle Basis Forecasts By Dhuyvetter, Kevin C.; Swanser, Kole; Kastens, Terry; Mintert, James; Crosby, Brett
  4. Quarterly Earnings Estimates for Publicly Traded Agribusinesses: An Evaluation By Manfredo, Mark; Sanders, Dwight; Scott, Winifred
  5. Regime switching models of hedge fund returns By Szabolcs Blazsek; Anna Downarowitz
  6. Large Bayesian VARs. By Marta Bańbura; Domenico Giannone; Lucrezia Reichlin
  7. Weather, Technology, and Corn and Soybean Yields in the U.S. Corn Belt By Tannura, Michael A.; Irwin, Scott H.; Good, Darrel L.
  8. Optimizing Time-series Forecasts for Inflation and Interest Rates Using Simulation and Model Averaging By Jumah, Adusei; Kunst, Robert M.

  1. By: Nazaria Solferino (Faculty of Economics, University of Rome "Tor Vergata"); Robert J. Waldmann (Faculty of Economics, University of Rome "Tor Vergata")
    Abstract: The signs of forecast errors can be predicted using the difference between individuals' forecasts and the average of earlier forecasts of the same variable. It is possible to improve forecasts without worsening any. It is difficult to reconcile this result with the rational expectations hypothesis, because the average of earlier forecasts is in the information set of the forecasters
    Keywords: Rational Expectations, Panel, Loss Function, Forecast, Interest Rate.
    JEL: G14 E47
    Date: 2008–11–24
  2. By: Mikko Kepsu; Hannu Schadewitz; Markku Vieru
    Abstract: ABSTRACT : Financial analysts comprise one important group of information intermediaries between firms and investors (Healy & Palepu, 2001). They have great potential to decrease information asymmetry between firms and investors, resulting in better allocation of capital. Analysts’ work is influenced by, among other things, the quality and quantity of information available from the target firms. Furthermore, analysts´ incentives could be influenced by the employer´s other affairs with the client. Our paper has three purposes : 1) to review the main research literature on analysts´ activity and performance, 2) to describe the development of analysts´ activity in the period 1987-2005 in a Finnish emerging market, and 3) to analyse the impact of market regulation and market cycles on analysts´ performance. Performance is studied in three dimensions : forecasting accuracy, forecast bias, and forecasting efficiency. Analysts´ data are based on I/B/E/S. Our analysis shows the rapid development of analysts´ activity, both in terms of the number of forecasts and longer forecasting horizons. Overall, the result supports the conclusion that analysts tend to be somewhat pessimistic in their Earnings per share (EPS) forecasts. Furthermore, the corrective actions taken have been somewhat sluggish (delays in EPS revisions). However, the forecasts improved significantly in the close before the actual EPS releases (0-1 month sample). Finally, analysts were not fully taking into account prior EPS development. This further supports the view that analysts underestimate the value of prior earnings change in their current earnings forecasting.
    Keywords: analysts´ earnings forecasting, emerging markets
    Date: 2008–11–19
  3. By: Dhuyvetter, Kevin C.; Swanser, Kole; Kastens, Terry; Mintert, James; Crosby, Brett
    Abstract: Forecasting feeder cattle basis has long been difficult because of the myriad factors that influence basis, including input and output prices and lot characteristics. This research draws upon knowledge of the various factors that influence cash feeder cattle prices to develop hedonic feeder cattle basis models. Out-of-sample test results provide strong evidence that these hedonic models predict basis more accurately than the multi-year average forecasting approach commonly used by livestock producers. Results from this research were used to develop a web tool funded by USDA's Risk Management Agency ( that producers can use to forecast and understand feeder cattle basis.
    Keywords: basis, basis forecasts, cattle prices, feeder cattle, hedging, price risk management, Agricultural Finance, Demand and Price Analysis, Farm Management, Livestock Production/Industries, Marketing, Risk and Uncertainty,
    Date: 2008–06–27
  4. By: Manfredo, Mark; Sanders, Dwight; Scott, Winifred
    Abstract: Decisions made by publicly traded agribusinesses impact suppliers, processors, farmers, and even rural communities. Professional analysts€٠estimates of earnings per share (EPS) provide a unique source of information regarding firm-level financial performance. Incorporating a battery of tests, this research examines the forecast properties of consensus analysts€٠EPS estimates reported in the Institutional Brokers Estimate System for a sample of publicly traded food companies. While the results are mixed among firms, they suggest 1) analysts forecasts are largely unbiased but inefficient, and may not encompass information in simple time series models, and 2) EPS may be becoming more difficult to estimate.
    Keywords: Earnings per share, forecasting, forecast evaluation, Agribusiness, Agricultural Finance, Financial Economics,
    Date: 2008–07
  5. By: Szabolcs Blazsek (Department of Business, Universidad de Navarra); Anna Downarowitz (Instituto de Empresa)
    Abstract: We estimate and compare the forecasting performance of several dynamic models of returns of different hedge fund strategies. The conditional mean of return is an ARMA process while its conditional volatility is modeled according to the GARCH specification. In order to take into account the high level of risk of these strategies, we also consider a Markov switching structure of the parameters in both equations to cap ture jumps. Finally, the one-step-ahead out-of-sample forecast performance of different models is compared.
    Keywords: Markov switching ARMA-GARCH, forecasting performance
    JEL: C13 C15 G32
    Date: 2008–11–26
  6. By: Marta Bańbura (European Central Bank, Kaiserstrasse 29, 60311 Frankfurt am Main, Germany.); Domenico Giannone (European Central Bank, Kaiserstrasse 29, 60311 Frankfurt am Main, Germany.); Lucrezia Reichlin (London Business School, Regents Park, London NW1 4SA, United Kingdom.)
    Abstract: This paper shows that Vector Autoregression with Bayesian shrinkage is an appropriate tool for large dynamic models. We build on the results by De Mol, Giannone, and Reichlin (2008) and show that, when the degree of shrinkage is set in relation to the cross-sectional dimension, the forecasting performance of small monetary VARs can be improved by adding additional macroeconomic variables and sectoral information. In addition, we show that large VARs with shrinkage produce credible impulse responses and are suitable for structural analysis. JEL Classification: C11, C13, C33, C53.
    Keywords: Bayesian VAR, Forecasting, Monetary VAR, large cross-sections.
    Date: 2008–11
  7. By: Tannura, Michael A.; Irwin, Scott H.; Good, Darrel L.
    Abstract: The purpose of this study was to investigate the relationship between weather, technology, and corn and soybean yields in the U.S. Corn Belt. Corn and soybean yields, monthly temperature, and monthly precipitation observations were collected over 1960 through 2006 for Illinois, Indiana, and Iowa. Multiple regression models were developed based on specifications found in studies by Thompson (1962 1963 1969 1970 1985 1986 1988). Estimated models explained at least 94% and 89% of the variation in corn and soybean yields for each state, respectively. Analysis of the regression results showed that corn yields were particularly affected by technology, the magnitude of precipitation during June and July, and the magnitude of temperatures during July and August. The effect of temperatures during May and June appeared to be minimal. Soybean yields were most affected by technology and the magnitude of precipitation during June through August (and especially during August). Structural change tests were performed on each model to test for changes in any of the regression model parameters. Some breakpoints were identified, but were difficult to explain since the results were not consistent across states and crops. Additional tests for structural change were directed specifically at the trend variable in corn models. The tests did not indicate a notable change in the technology trend for corn since the mid-1990s. Corn and soybean yield forecasts from the regression models on June 1 and July 1 were no more accurate then trend yield forecasts. Regression model forecasts for corn improved on August 1, while model forecasts for soybeans improved by September 1. U.S. Department of Agriculture (USDA) corn and soybean forecasts were always more accurate than those from the regression models. Nonetheless, encompassing tests showed that the accuracy of USDA yield forecasts could be significantly improved by the information contained in regression model forecasts. Across states and forecast months, combining regression model forecasts with USDA forecasts improved accuracy an average of 10% for corn and 6% for soybeans. In sum, this research provided strong evidence that precipitation, temperature, and a linear time trend to represent technological improvement explained all but a small portion of the variation in corn and soybean yields in the U.S. Corn Belt. An especially important finding was that relatively benign weather for the development of corn since the mid-1990s should not be discounted as an explanation for seemingly €ܨigh€ݠyields. The potential impact of this finding on the agricultural sector is noteworthy. Trend yield forecasts based on perceptions of a rapid increase in technology may eventually lead to poor forecasts. Unfavorable weather in the future may lead to unexpectedly low corn yields that leave producers, market participants, and policymakers wondering how such low yields could have occurred despite technological improvements.
    Keywords: Agricultural Finance, Financial Economics, Research Methods/ Statistical Methods,
    Date: 2008–02
  8. By: Jumah, Adusei (Department of Economics and Finance, Institute for Advanced Studies, Vienna, Austria, and Department of Economics, University of Vienna, Vienna, Austria); Kunst, Robert M. (Department of Economics and Finance, Institute for Advanced Studies, Vienna, Austria, and Department of Economics, University of Vienna, Vienna, Austria)
    Abstract: Motivated by economic-theory concepts—the Fisher hypothesis and the theory of the term structure—we consider a small set of simple bivariate closed-loop time-series models for the prediction of price inflation and of long- and short-term interest rates. The set includes vector autoregressions (VAR) in levels and in differences, a cointegrated VAR, and a non-linear VAR with threshold cointegration based on data from Germany, Japan, UK, and the U.S. Following a traditional comparative evaluation of predictive accuracy, we subject all structures to a mutual validation using parametric bootstrapping. Ultimately, we utilize the recently developed technique of Mallows model averaging to explore the potential of improving upon the predictions through combinations. While the simulations confirm the traded wisdom that VARs in differences optimize one-step prediction and that error correction helps at larger horizons, the model-averaging experiments point at problems in allotting an adequate penalty for the complexity of candidate models.
    Keywords: Threshold cointegration, Parametric bootstrap, Model averaging
    JEL: C32 C52 E43 E47
    Date: 2008–11

This nep-for issue is ©2008 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.