nep-for New Economics Papers
on Forecasting
Issue of 2014‒01‒24
five papers chosen by
Rob J Hyndman
Monash University

  1. Benchmarking time series based forecasting models for electricity balancing market prices By Gro Klaeboe; Anders Lund Eriksrud; Stein-Erik Fleten
  2. Rainfall Forecasts, Weather and Wages over the Agricultural Production Cycle By Mark R. Rosenzweig; Christopher Udry
  3. Simultaneous Confidence Corridors and Variable Selection for Generalized Additive Models By Shuzhuan Zheng; Rong Liu; Lijian Yang; Wolfgang Karl Härdle
  4. A consistent two-factor model for pricing temperature derivatives By Andreas Groll; Brenda López-Cabrera; Thilo Meyer-Brandis;
  5. Confidence Bands for Impulse Responses: Bonferroni versus Wald By Helmut Lütkepohl; Anna Staszewska-Bystrova; Peter Winker;

  1. By: Gro Klaeboe (Norwegian University of Science and Technology); Anders Lund Eriksrud (Norwegian University of Science and Technology); Stein-Erik Fleten (Norwegian University of Science and Technology)
    Abstract: In the trade-off between bidding in the day-ahead electricity market and the real time balancing market, producers need good forecasts for balancing market prices to make informed decisions. A range of earlier published models for forecasting of balancing market prices, including a few extensions, is benchmarked. The models are benchmarked both for one hour-ahead and day-ahead forecast, and both point and interval forecasts are compared. None of the benchmarked models produce infor- mative day-ahead point forecasts, suggesting that information available before the closing of the day-ahead market is eciently re ected in the day-ahead market price rather than the balancing market price. Evalua- tion of the interval forecasts reveals that models without balancing state information overestimate variance, making them unsuitable for scenario generation.
    Keywords: Federal Reserve, Forecast Evaluation, Survey of Professional Forecasts, Business Cycle, Mahalanobis Distance
    JEL: C5 E2 E3
    Date: 2013–10
  2. By: Mark R. Rosenzweig; Christopher Udry
    Abstract: We look at the effects of rainfall forecasts and realized rainfall on equilibrium agricultural wages over the course of the agricultural production cycle. We show theoretically that a forecast of good weather can lower wages in the planting stage, by lowering ex ante out-migration, and can exacerbate the negative impact of adverse weather on harvest-stage wages. Using Indian household panel data describing early-season migration and district-level planting- and harvest-stage wages over the period 2005-2010, we find results consistent with the model, indicating that rainfall forecasts improve labor allocations on average but exacerbate wage volatility because they are imperfect.
    JEL: J2 J31 J43 O1 O13 Q12
    Date: 2014–01
  3. By: Shuzhuan Zheng; Rong Liu; Lijian Yang; Wolfgang Karl Härdle
    Abstract: In spite of the widespread use of generalized additive models (GAMs), there is no well established methodology for simultaneous inference and variable selection for the components of GAM. There is no doubt that both, inference on the marginal component functions and their selection, are essential in this additive statistical models. To this end, we establish simultaneous confidence corridors (SCCs) and a variable selection criteria through the spline-backfitted kernel smoothing techniques. To characterize the global features of each component, SCCs are constructed for testing their shapes. By extending the BIC to additive models with identity/trivial link, an asymptotically consistent BIC approach for variable selection is proposed. Our procedures are examined in simulations for its theoretical accuracy and performance, and used to forecast the default probability of listed Japanese companies.
    Keywords: BIC, Confidence corridor, Extreme value, Generalized additive model, Spline-backfitted kernel
    JEL: C35 C52 C53 G33
    Date: 2014–01
  4. By: Andreas Groll; Brenda López-Cabrera; Thilo Meyer-Brandis;
    Abstract: We analyze a consistent two-factor model for pricing temperature derivatives that incorporates the forward looking information available in the market by specifying a model for the dynamics of the complete meteorological forecast curve. The two-factor model is a generalization of the Nelson-Siegel curve model by allowing factors with mean-reversion to a stochastic mean for structural changes and seasonality for periodic patterns. Based on the outcomes of a statistical analysis of forecast data we conclude that the two-factor model captures well the stylized features of temperature forecast curves. In particular, a functional principal component analysis reveals that the model re ects reasonably well the dynamical structure of forecast curves by decomposing their shapes into a tilting and a bending factor. We continue by developing an estimation procedure for the model, before we derive explicit prices for temperature derivatives and calibrate the market price of risk (MPR) from temperature futures derivatives (CAT, HDD, CDD) traded at the Chicago Mercantile Exchange (CME). The factor model shows that the behavior of the implied MPR for futures traded in and out of the measurement period is more stable than other estimates obtained in the literature. This conrms that at least parts of the irregularity of the MPR is not due to irregular risk perception but rather due to information misspecification. Similar to temperature derivatives, this approach can be used for pricing other non-tradable assets.
    Keywords: factor models, consistency, pricing and hedging, weather derivatives, market price of risk
    JEL: G19 G29 G22 N23 N53 Q59
    Date: 2014–01
  5. By: Helmut Lütkepohl; Anna Staszewska-Bystrova; Peter Winker;
    Abstract: In impulse response analysis estimation uncertainty is typically displayed by constructing bands around estimated impulse response functions. These bands may be based on frequentist or Bayesian methods. If they are based on the joint distribution in the Bayesian framework or the joint asymptotic distribution possibly constructed with bootstrap methods in the frequentist framework often individual confidence intervals or credibility sets are simply connected to obtain the bands. Such bands are known to be too narrow and have a joint confidence content lower than the desired one. If instead the joint distribution of the impulse response coefficients is taken into account and mapped into the band it is shown that such a band is typically rather conservative. It is argued that a smaller band can often be obtained by using the Bonferroni method. While these considerations are equally important for constructing forecast bands, we focus on the case of impulse responses in this study.
    Keywords: Impulse responses, Bayesian error bands, frequentist confidence bands, Wald statistic, vector autoregressive process
    JEL: C32
    Date: 2014–01

This nep-for issue is ©2014 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.