In this issue we have: Does forecast combination improve Norges Bank inflation forecasts? ; Understanding forecast failure in ESTAR models of real exchange rates ; The Taylor rule and forecast intervals for exchange rates ; Risk-adjusted forecasts of oil prices ; and more.

- Does forecast combination improve Norges Bank inflation forecasts?

Date: 2009-01-27 By: Hilde C. Bjørnland (Norwegian School of Management (BI) and Norges Bank (Central Bank of Norway))

Karsten Gerdrup (Norges Bank (Central Bank of Norway))

Anne Sofie Jore (Norges Bank (Central Bank of Norway))

Christie Smith (Norges Bank (Central Bank of Norway))

Leif Anders Thorsrud (Norges Bank (Central Bank of Norway))URL: http://d.repec.org/n?u=RePEc:bno:worpap:2009_01&r=for We develop a system that provides model-based forecasts for inflation in Norway. Forecasts are recursively evaluated from 1999 to 2008. The performance of the models over this period is then used to derive weights that are used to combine the forecasts. Our results indicate that model combination improves upon the point forecasts from individual models. Furthermore, when comparing the whole forecasting period; model combination outperforms Norges Banks own point forecast for inflation at the forecast horizon up to a year. By using a suite of models we allow for a greater range of modelling techniques and data to be used in the forecasting process. Keywords: Forecasting, forecast combination JEL: E52 E37 E47 - Understanding forecast failure in ESTAR models of real exchange rates

Date: 2009-02-03 By: Buncic, Daniel URL: http://d.repec.org/n?u=RePEc:pra:mprapa:13121&r=for The forecast performance of the empirical ESTAR model of Taylor et al. (2001) is examined for 4 bilateral real exchange rate series over an out-of-sample evaluation period of nearly 12 years. Point as well as density forecasts are evaluated relative to a simple AR(1) specification, considering horizons up to 22 steps head. The results of this study suggest that no forecast gains over a simple AR(1) model exist at any of the forecast horizons that are considered, regardless of whether point or density forecasts are used. Using simulation and non-parametric techniques in conjunction with graphical methods, this study shows that the non-linearity in the point forecasts of the ESTAR model decrease as the forecast horizon increases. Multiple steps ahead density forecasts of the ESTAR model are approximately normal looking, with no signs of skewness or bimodality. For an applied forecaster, there do not appear to exist any gai! ns in using the non-linear ESTAR model over a simple AR(1) specification. Keywords: Purchasing power parity; regime modelling; non-linear real exchange rate models; ESTAR; forecast evaluation; density forecasts; non-parametric methods. JEL: C53 C52 C22 F47 F31 - The Taylor rule and forecast intervals for exchange rates

Date: 2009 By: Jian Wang

Jason J. WuURL: http://d.repec.org/n?u=RePEc:fip:fedgif:963&r=for This paper attacks the Meese-Rogoff (exchange rate disconnect) puzzle from a different perspective: out-of-sample interval forecasting. Most studies in the literature focus on point forecasts. In this paper, we apply Robust Semi-parametric (RS) interval forecasting to a group of Taylor rule models. Forecast intervals for twelve OECD exchange rates are generated and modified tests of Giacomini and White (2006) are conducted to compare the performance of Taylor rule models and the random walk. Our contribution is twofold. First, we find that in general, Taylor rule models generate tighter forecast intervals than the random walk, given that their intervals cover out-of-sample exchange rate realizations equally well. This result is more pronounced at longer horizons. Our results suggest a connection between exchange rates and economic fundamentals: economic variables contain information useful in forecasting the distribution! s of exchange rates. The benchmark Taylor rule model is also found to perform better than the monetary and PPP models. Second, the inference framework proposed in this paper for forecast-interval evaluation, can be applied in a broader context, such as inflation forecasting, not just to the models and interval forecasting methods used in this paper. - Risk-adjusted forecasts of oil prices

Date: 2009-01 By: Patrizio Pagano (Bank of Italy, via Nazionale 91, I – 00184 Rome, Italy.)

Massimiliano Pisani (Bank of Italy, via Nazionale 91, I – 00184 Rome, Italy.)URL: http://d.repec.org/n?u=RePEc:ecb:ecbwps:20090999&r=for This paper documents the existence of a significant forecast error on crude oil futures. We interpret it as a risk premium, which, in part, could have been explained by means of a real-time US business cycle indicator, such as the degree of capacity utilization in manufacturing. This result is robust to the specification of the estimating equation and to the considered business cycle indicator. An out-of-the-sample prediction exercise reveals that futures adjusted to take into account this time-varying component produce significantly better forecasts than those of unadjusted futures, of futures adjusted for the average forecast error and of the random walk, particularly at horizons of more than 6 months. JEL Classification: E37, E44, G13, Q4. Keywords: Oil, Forecasting, Futures. - Bootstrap prediction intervals for threshold autoregressive models

Date: 2009-01 By: Jing, Li URL: http://d.repec.org/n?u=RePEc:pra:mprapa:13086&r=for This paper examines the performance of prediction intervals based on bootstrap for threshold autoregressive models. We consider four bootstrap methods to account for the variability of estimates, correct the small-sample bias of autoregressive coefficients and allow for heterogeneous errors. Simulation shows that (1) accounting for the sampling variability of estimated threshold values is necessary despite super-consistency, (2) bias-correction leads to better prediction intervals under certain circumstances, and (3) two-sample bootstrap can improve long term forecast when errors are regime-dependent. Keywords: Bootstrap; Interval Forecasting; Threshold Autoregressive Models; Time Series; Simulation JEL: C53 C22 C15 - Path Forecast Evaluation

Date: 2008-07 By: Jorda, Oscar (U of California, Davis)

Marcellino, Massimiliano (Universita Bocconi)URL: http://d.repec.org/n?u=RePEc:ecl:ucdeco:08-5&r=for A path forecast refers to the sequence of forecasts 1 to H periods into the future. A summary of the range of possible paths the predicted variable may follow for a given confidence level requires construction of simultaneous confidence regions that adjust for any covariance between the elements of the path forecast. This paper shows how to construct such regions with the joint predictive density and Scheffe's (1953) S-method. In addition, the joint predictive density can be used to construct simple statistics to evaluate the local internal consistency of a forecasting exercise of a system of variables. Monte Carlo simulations demonstrate that these simultaneous confidence regions provide approximately correct coverage in situations where traditional error bands, based on the collection of marginal predictive densities for each horizon, are vastly off mark. The paper showcases these methods with an application to the mos! t recent monetary episode of interest rate hikes in the U.S. macroeconomy. JEL: C32 - Forecasting Demand for Rural Electric Cooperative Call Center

Date: 2009-01 By: Kim, Taeyoon

Kenkel, Philip

Brorsen, B. WadeURL: http://d.repec.org/n?u=RePEc:ags:saeana:46809&r=for This research forecasts peak call volume to allow a centralized call center to minimize staffing costs. A Gaussian copula is used to capture the dependence among nonnormal distributions. Peak call volume can be easily and more accurately predicted using the marginal probability distribution with the copula function than without using a copula. The modeling approach allows simulating adding another cooperative. Ignoring the dependence that the copula includes, causes peak values to be underestimated. Keywords: Call center data, Empirical distibution, Extreme value theory, Forecasting, Gamma distribution, Gaussian copula, Simulation, Agribusiness, Demand and Price Analysis, Research Methods/ Statistical Methods, - Estimating components of ICT expenditure: a model-based approach with applicability to short time-series

Date: 2008 By: Cooper, Russel

Madden, Gary GURL: http://d.repec.org/n?u=RePEc:pra:mprapa:13007&r=for This paper develops a microeconomic model-based approach to forecast national information and communications technology expenditure that is helpful when only very short time-series are available. The model specification incorporates parameters for network effects and national e-readiness. Finally, the model allows for observed non-homotheticity and ‘noise' found in sample data, with the latter attributed to country-specific influences. Keywords: ICT forecasts; short time-series; microeconomic modeling JEL: C51 L96 C53 - Predicting the Corn Basis in the Texas Triangle Area

Date: 2009-01-15 By: Mkrtchyan, Vardan

Welch, J. Mark

Power, Gabriel J.URL: http://d.repec.org/n?u=RePEc:ags:saeana:46759&r=for This study develops a new and straightforward economic model of basis forecasting that outperforms the simple three-year average method suggested in much of the literature. We use monthly data of the corn basis in the Texas Triangle Area from February 1997 to July 2008. The results and the graphs indicate that the new model based on economic fundamentals performs better than basis estimates using a three-year moving average. Keywords: Hedging, basis, corn, Agribusiness, Agricultural Finance, Financial Economics, Marketing, Risk and Uncertainty, - Land Use Change and Ecosystem Valuation in North Georgia

Date: 2009-01 By: Ngugi, D

Mullen, J

Bergstrom, JURL: http://d.repec.org/n?u=RePEc:ags:saeana:46853&r=for This study seeks to forecast land use change in a North Georgia ecosystem, and estimate the economic value of the ecosystem using benefit transfer techniques. We forecast land use change based on a structural time series model and a simple growth rate model. The study suggests a lower bound willingness to pay value of about USD 16,000 per year to ensure compliance with fishing and drinking water quality standards with regard to fecal coliform bacteria and dissolved oxygen. Conservation efforts are likely to cost less than the cost of defensive behavior or ecosystem restoration. Keywords: Ecosystem, Economic value, North Georgia, land use, water quality, structural time series, benefit transfer, forecasting., Environmental Economics and Policy, Land Economics/Use, Q51, Q53, Q57, - Sunshine-Factor Model with Treshold GARCH for Predicting Temperature of Weather Contracts

Date: 2008-08 By: Hélène Hamisultane (EconomiX – CNRS : UMR7166 – Université de Paris X – Nanterre) URL: http://d.repec.org/n?u=RePEc:hal:wpaper:halshs-00355857_v1&r=for Climate changes have sparked growing interest for the weather derivatives which are financial contracts relied on a meteorological index and allowing companies to hedge against climate risk. These contracts present the particularity of providing compensation to the buyer when the meteorological index crossed a limit agreed in advance with the seller. In order to evaluate these products and to manage at best the risks associated with their exchange, it is important to be able to accurately predict the evolution of the climate variable. Several processes have been proposed in the literature to model the behaviour of the temperature which is the basis of most of the traded weather instruments. These processes relate mainly to the univariate time series modelling which is founded on the study of the autocorrelation of the stationary variable. But we know that the behaviour of the temperature can be influenced by climatic fac! tors such as rain, wind or sunshine. In our paper, we propose to take into account the impact of sunshine on the temperature as well as the asymmetric effect of the shocks on the volatility by estimating a structural model with a periodic threshold GARCH. We show that this model provides better out-sample forecasts for 30 and 60 days ahead than those obtained by the univariate autoregressive-conditional heteroskedasticity process. Keywords: weather derivatives; structural model; Markov chain; threshold GARCH; Monte-Carlo simulations; Value-at-Risk. - Estimating Term Structure Equations Using Macroeconomic Variables

Date: 2008-01 By: Fair, Ray C. (Yale U) URL: http://d.repec.org/n?u=RePEc:ecl:yaleco:32&r=for This paper begins with the expectations theory of the term structure of interest rates with constant term premia and then postulates how expectations of future short term interest rates are formed. Expectations depend in part on predictions from a set of VAR equations and in part on the current and two lagged values of the short term interest rate. The results suggest that there is relevant independent information in both the VAR equations' predictions and the current and two lagged values of the short rate. The model fits the long term interest rate data well, including the 2004-2006 period, which some have found a puzzle. The properties of the model are consistent with the response of the long term U.S. Treasury bond rate to surprise price and employment announcements. The overall results suggest that long term rates can be fairly well explained by modeling expectation formation of future short term rates. JEL: E43

Taken from the NEP-FOR mailing list edited by Rob Hyndman.