nep-for New Economics Papers
on Forecasting
Issue of 2010‒04‒24
six papers chosen by
Rob J Hyndman
Monash University

  1. Forecasting Nonlinear Aggregates and Aggregates with Time-varying Weights By Helmut Luetkepohl
  2. Forecasting Government Bond Yields with Large Bayesian VARs By Andrea Carriero; George Kapetanios; Massimiliano Marcellino
  3. New Keynesian Model Features that Can Reproduce Lead, Lag and Persistence Patterns. By Steven P. Cassou; Jesús Vázquez
  4. How Risky Is the Value at Risk? By Roxana Chiriac; Winfried Pohlmeier
  5. A note on GDP now-/forecasting with dynamic versus static factor models along a business cycle By Buss, Ginters
  6. Optimal predictions of powers of conditionally heteroskedastic processes By Francq, Christian; Zakoian, Jean-Michel

  1. By: Helmut Luetkepohl
    Abstract: Despite the fact that many aggregates are nonlinear functions and the aggregation weights of many macroeconomic aggregates are timevarying, much of the literature on forecasting aggregates considers the case of linear aggregates with fixed, time-invariant aggregation weights. In this study a framework for nonlinear contemporaneous aggregation with possibly stochastic or time-varying weights is developed and different predictors for an aggregate are compared theoretically as well as with simulations. Two examples based on European unemployment and inflation series are used to illustrate the virtue of the theoretical setup and the forecasting results.
    Keywords: Forecasting, stochastic aggregation, autoregression, moving average,vector autoregressive process
    JEL: C32
    Date: 2010
  2. By: Andrea Carriero (Queen Mary, University of London); George Kapetanios (Queen Mary, University of London); Massimiliano Marcellino (European University Institute and Bocconi University)
    Abstract: We propose a new approach to forecasting the term structure of interest rates, which allows to efficiently extract the information contained in a large panel of yields. In particular, we use a large Bayesian Vector Autoregression (BVAR) with an optimal amount of shrinkage towards univariate AR models. Focusing on the U.S., we provide an extensive study on the forecasting performance of our proposed model relative to most of the existing alternative specifications. While most of the existing evidence focuses on statistical measures of forecast accuracy, we also evaluate the performance of the alternative forecasts when used within trading schemes or as a basis for portfolio allocation. We extensively check the robustness of our results via subsample analysis and via a data based Monte Carlo simulation. We find that: i) our proposed BVAR approach produces forecasts systematically more accurate than the random walk forecasts, though the gains are small; ii) some models beat the BVAR for a few selected maturities and forecast horizons, but they perform much worse than the BVAR in the remaining cases; iii) predictive gains with respect to the random walk have decreased over time; iv) different loss functions (i.e., "statistical" vs "economic") lead to different ranking of specific models; v) modelling time variation in term premia is important and useful for forecasting.
    Keywords: Bayesian methods, Forecasting, Term structure
    JEL: C11 C53 E43 E47
    Date: 2010–04
  3. By: Steven P. Cassou (Kansas State University); Jesús Vázquez (Universidad del País Vasco)
    Abstract: This paper uses a new method for describing dynamic comovement and persistence in economic time series which builds on the contemporaneous forecast error method developed in den Haan (2000). This data description method is then used to address issues in New Keynesian model performance in two ways. First, well known data patterns, such as output and inflation leads and lags and inflation persistence, are decomposed into forecast horizon components to give a more complete description of the data patterns. These results show that the well known lead and lag patterns between output and inflation arise mostly in the medium term forecasts horizons. Second, the data summary method is used to investigate a rich New Keynesian model with many modeling features to see which of these features can reproduce lead, lag and persistence patterns seen in the data. Many studies have suggested that a backward looking component in the Phillips curve is needed to match the data, but our simulations show this is not necessary. We show that a simple general equilibrium model with persistent IS curve shocks and persistent supply shocks can reproduce the lead, lag and persistence patterns seen in the data.
    Keywords: New Keynesian, output and inflation comovement, inflation persistence, forecast error
    JEL: E31 E32 E37
    Date: 2010–04–15
  4. By: Roxana Chiriac (University of Konstanz, CoFE); Winfried Pohlmeier (University of Konstanz, CoFE, ZEW, RCEA)
    Abstract: The recent financial crisis has raised numerous questions about the accuracy of value-at-risk (VaR) as a tool to quantify extreme losses. In this paper we present empirical evidence from assessing the out-of-sample performance and robustness of VaR before and during the recent financial crisis with respect to the choice of sampling window, return distributional assumptions and stochastic properties of the underlying financial assets. Moreover we develop a new data driven approach that is based on the principle of optimal combination and that provides robust and precise VaR forecasts for periods when they are needed most, such as the recent financial crisis.
    Keywords: Value at Risk, model risk, optimal forecast combination
    JEL: C21 C5 G28 G32
    Date: 2010–01
  5. By: Buss, Ginters
    Abstract: We build a small-scale factor model for the GDP of one of the hardest hit economies during the latest recession to study the exact dynamic versus static factor model performance along a business cycle, with an emphasis placing on nowcasting performance during a pronounced switch of business cycle phases due to the latest recession. We compare the factor models' nowcasting performance to a random walk, autoregressive and the best-performing nowcasting models at our hands, which are vector autoregressive (VAR) models. It is shown that a small-scale static factor-augmented VAR (FAVAR) model tends to improve upon the nowcasting performance of the VAR models when the model span and the nowcasting period stretch beyond a single business cycle phase, while exact dynamic factor models tend to fail to detect the timing and depth of the recession regardless of ARMA specifications. As regards the case when the model span and the nowcasting period are contained within a single business cycle phase, static and dynamic factor models appear to show similar performance with potentially slight superiority of dynamic factor models if the factor-forming set of variables and factor dynamics are carefully selected.
    Keywords: nowcasting; business cycle; static versus dynamic factors; small-scale FAVAR; VAR; GDP
    JEL: C32 C53 C52 C22
    Date: 2010–04–16
  6. By: Francq, Christian; Zakoian, Jean-Michel
    Abstract: In conditionally heteroskedastic models, the optimal prediction of powers, or logarithms, of the absolute process has a simple expression in terms of the volatility process and an expectation involving the independent process. A standard procedure for estimating this prediction is to estimate the volatility by gaussian quasi-maximum likelihood (QML) in a first step, and to use empirical means based on rescaled innovations to estimate the expectation in a second step. This paper proposes an alternative one-step procedure, based on an appropriate non-gaussian QML estimation of the model, and establishes the asymptotic properties of the two approaches. Their performances are compared for finite-order GARCH models and for the infinite ARCH. For the standard GARCH(p, q) and the Asymmetric Power GARCH(p,q), it is shown that the ARE of the estimators only depends on the prediction problem and some moments of the independent process. An application to indexes of major stock exchanges is proposed.
    Keywords: APARCH; Infinite ARCH; Conditional Heteroskedasticity; Efficiency of estimators; GARCH; Prediction; Quasi Maximum Likelihood Estimation
    JEL: C13 C22 C01
    Date: 2010–04–17

This nep-for issue is ©2010 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.