nep-for New Economics Papers
on Forecasting
Issue of 2007‒03‒24
six papers chosen by
Rob J Hyndman
Monash University

  1. Forecasts of US Short-term Interest Rates: A Flexible Forecast Combination Approach By Guidolin, Massimo; Timmermann, Allan G
  2. Heavy tails and electricity prices: Do time series models with non-Gaussian noise forecast better? By Weron, Rafal; Misiorek, Adam
  4. Is core money growth a good and stable inflation predictor in the euro area? By Kai Carstensen
  5. Notional Defined Contribution Accounts (NDCs): Solvency and Risk; Application to the Case of Spain By Carlos Vidal-Meliá; Inmaculada Domínguez-Fabián; María del Carmen Boado-Penas
  6. Area Based Models of New Highway Route Growth By David Levinson; Wei Chen

  1. By: Guidolin, Massimo; Timmermann, Allan G
    Abstract: This paper develops a flexible approach to combine forecasts of future spot rates with forecasts from time-series models or macroeconomic variables. We find empirical evidence that accounting for both regimes in interest rate dynamics and combining forecasts from different models helps improve the out-of-sample forecasting performance for US short-term rates. Imposing restrictions from the expectations hypothesis on the forecasting model are found to help at long forecasting horizons.
    Keywords: forecast combinations; term structure of interest rates
    JEL: C53 G12
    Date: 2007–03
  2. By: Weron, Rafal; Misiorek, Adam
    Abstract: This paper is a continuation of our earlier studies on short-term price forecasting of California electricity prices with time series models. Here we focus on whether models with heavy-tailed innovations perform better in terms of forecasting accuracy than their Gaussian counterparts. Consequently, we limit the range of analyzed models to autoregressive time series approaches that have been found to perform well for pre-crash California power market data. We expand them by allowing for heavy-tailed innovations in the form of α-stable or generalized hyperbolic noise.
    Keywords: Electricity; price forecasting; heavy tails; time series; α-stable distribution; generalized hyperbolic distribution
    JEL: C53 C46 C22 Q40
    Date: 2007–03
  3. By: Brian S. Fisher; Guy Jakeman; Hom M. Pant; Malte Schwoon; Richard S.J. Tol (Economic and Social Research Institute, Dublin)
    Abstract: We present the Canberra-Hamburg Integrated Model for Population (CHIMP), a new global population model for long-term projections. Distinguishing features of this model, compared to other model for secular population projections, are that (a) mortality, fertility, and migration are partly driven by per capita income; (b) large parts of the model have been estimated rather than calibrated; and (c) the model is in the public domain. Scenario experiments show similarities but also differences with other models. Similarities include rapid aging of the population and an eventual reversal of global population growth. The main difference is that CHIMP projects substantially higher populations, particularly in Africa, primarily because our data indicate a slower fertility decline than assumed elsewhere. Model runs show a strong interaction between population growth and economic growth, and a weak feedback of climate change on population growth.
    Keywords: population model, long term projections, global change, integrated assessment
    JEL: J11 Q54
    Date: 2005–07
  4. By: Kai Carstensen
    Abstract: In this paper, it is analyzed whether core money growth helps to predict future inflation in a useful and reliable way. Using an out-of-sample forecasting exercise and a stability analysis, it is shown that core money growth carries important information not contained in the inflation history, that its inclusion in a forecasting model can increase the forecasting accuracy, and that it has had a strong and stable long-run link to inflation over the last decades. A particularly promising forecasting model at all horizons is the one proposed by Gerlach (2004) that includes the inflation gap, the difference between core money growth and core inflation, and the output gap. This model has a very good track record, exhibits stable parameters over both the pre-EMU and the EMU era. What makes it appealing from a more theoretical perspective is that it relies on the stable long-run relationship between money growth and inflation.
    Keywords: Forecasting, core money growth, stability, filter
    JEL: E47 E58
    Date: 2007–02
  5. By: Carlos Vidal-Meliá; Inmaculada Domínguez-Fabián; María del Carmen Boado-Penas
    Abstract: The aim of this work is twofold, on the one hand, to demonstrate the actuarial imbalance of the Spanish pension system in its current configuration, and on the other, to measure the aggregate economic risk to which the pensioner would be exposed if it were decided to apply ten formulas for the calculation of the retirement pension based on notional accounts. Given the uncertainty involved in working with a long term horizon, a model of generation of multi-periodic scenarios is used, based on the predictions of mean values of Alonso and Herce (2003) for the period 2006-2050. This provides up to ten thousand trajectories of the macroeconomic indices needed to calculate such parameters as the initial pension, the replacement rate (RR) or the internal rate of return (IRR), and the value-at-risk (VaR) of the pensioner. The results obtained are analyzed in both objective and subjective terms. The main conclusions are that, applying the notional philosophy, the expected average RR and IRR would be much lower than those obtained under the current rules of the pay-as-you-go system. If the projections used were slightly probable, the pension system would build up such a large additional financial imbalance in the future that it would require either a considerable reduction in the initial pension or a severe combination of parameter adjustments. From the risk perspective, the preferred formulas for a beneficiary most averse would be those based on future variations in salaries with a pension constant in real terms, whereas those beneficiaries less averse to risk would prefer formulas supplying a lower initial pension which grows in real terms in line with future variations in salaries.
  6. By: David Levinson (Nexus (Networks, Economics, and Urban Systems) Research Group, Department of Civil Engineering, University of Minnesota); Wei Chen
    Abstract: Empirical data and statistical models are used to answer the question of where new highway routes are most likely to be located. High-quality land-use, population distribution and highway network GIS data for the Twin CitiesMetropolitan Area from 1958 to 1990 are developed for this study. The highway system is classified into three levels, Interstate highways, divided highways, andsecondary highways. Binary logit models estimate the new route growth probability of divided highways and secondary highways. Interstates, however,are not modeled here and are used as a predictor in modeling the growth of divided highways and secondary highways. The results show that the area's land-use attributes and population density level do have significant relationship with the area's likelihood of adding new highway routes.
    Keywords: network growth, hierarchy of roads, land-use, population, GIS.
    JEL: R41 R42 R48 O33
    Date: 2007

This nep-for issue is ©2007 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.