nep-for New Economics Papers
on Forecasting
Issue of 2017‒02‒12
eight papers chosen by
Rob J Hyndman
Monash University

  1. Valuing predictability By Antony Millner; Daniel Heyen
  2. Bayesian Semiparametric Forecasts of Real Interest Rate Data By DESCHAMPS, Philippe J.
  3. How Biased Are U.S. Government Forecasts of the Federal Debt? By Neil R. Ericsson
  4. Nowcasting the Czech Trade Balance By Oxana Babecka Kucharcukova; Jan Bruha
  5. “Regional tourism demand forecasting with machine learning models: Gaussian process regression vs. neural network models in a multiple-input multiple-output setting" By Oscar Claveria; Enric Monte; Salvador Torra
  6. A New Approach to Volatility Modeling : The High-Dimensional Markov Model By AUGUSTYNIAK, Maciej; BAUWENS, Luc; DUFAYS, Arnaud
  7. Mixture Normal Conditional Correlation Models By Maria Putintseva
  8. Representation, Estimation and Forecasting of the Multivariate Index-Augmented Autoregressive Model By Gianluca Cubadda; Barbara Guardabascio

  1. By: Antony Millner; Daniel Heyen
    Abstract: How important is it to be able to predict the distant future? The authors of this paper study this question in a model of an agent who operates in a non-stationary stochastic environment. Payoffs depend on how well adapted activities are to current conditions, and activities may be adjusted to account for anticipated environmental changes, at a cost. The authors compute the value of prediction systems, which produce forecasts of the future with a given profile of accuracy as a function of lead time in every period. This allows them to quantify the importance of predictive accuracy at each lead time. Even if adjustment costs, discount factors and long-run uncertainty are large, short-run predictability is often more important than long-run predictability. ‘If you have to forecast, forecast often.’ Edgar R. Fiedler, The Three Rs of Economic Forecasting: Irrational, Irrelevant and Irreverent, 1977
    Date: 2017–01
  2. By: DESCHAMPS, Philippe J. (Université catholique de Louvain, CORE, Belgium)
    Abstract: The non-hierarchical Dirichlet process prior has been mainly used for parameters of innovation distributions. It is, however, easy to apply to all the parameters (coefficients of covariates and innovation variance) of more general regression models. This paper investigates the predictive performance of a simple (non-hierarchical) Dirichlet process mixture of Gaussian autoregressions for forecasting monthly US real interest rate data. The results suggest that the number of mixture components increases sharply over time, and the predictive marginal likelihoods strongly dominate those of a benchmark autoregressive model. Unconditional predictive coverage is vastly improved in the mixture model.
    Keywords: Dirichlet process mixture, Bayesian nonparametrics, structural change, real interest rate
    JEL: C11 C14 C22 C53
    Date: 2016–11–01
  3. By: Neil R. Ericsson
    Abstract: Government debt and forecasts thereof attracted considerable attention during the recent financial crisis. The current paper analyzes potential biases in different U.S. government agencies’ one-year-ahead forecasts of U.S. gross federal debt over 1984-2012. Standard tests typically fail to detect biases in these forecasts. However, impulse indicator saturation (IIS) detects economically large and highly significant time-varying biases, particularly at turning points in the business cycle. These biases do not appear to be politically related. IIS defines a generic procedure for examining forecast properties; it explains why standard tests fail to detect bias; and it provides a mechanism for potentially improving forecasts.
    Keywords: Autometrics ; Bias ; Debt ; Federal government ; Forecasts ; Impulse indicator saturation ; Heteroscedasticity ; Projections ; United States
    JEL: H68 C53
    Date: 2017–01–06
  4. By: Oxana Babecka Kucharcukova; Jan Bruha
    Abstract: In this paper we are interested in nowcasting and short-run forecasting of the main external trade variables. We consider four empirical methods: principal component regression, elastic net regression, the dynamic factor model and partial least squares. We discuss the adaptation of those methods to asynchronous data releases and to the mixed-frequency set-up. We contrast them with a set of univariate benchmarks. We find that for variables in value terms (both nominal and real), elastic net regression typically yields the most accurate predictions, followed by the dynamic factor model and then by principal components. For export and import prices, univariate techniques seem to have the higher precision for backcasting and nowcasting, but for short-run forecasting the more sophisticated methods tend to produce more accurate forecasts. Here again, elastic net regression dominates the other methods.
    Keywords: Dynamic factor models, elastic net regression, mixed-frequency data, nowcasting, principal component analysis, state space models, trade balance
    JEL: C53 C55 F17
    Date: 2016–12
  5. By: Oscar Claveria (AQR Research Group-IREA. University of Barcelona. Av.Diagonal 696; 08034 Barcelona, Spain.); Enric Monte (Department of Signal Theory and Communications, Polytechnic University of Catalunya (UPC).); Salvador Torra (Riskcenter-IREA, University of Barcelona, Av. Diagonal 690, 08034 Barcelona, Spain.)
    Abstract: This study presents a multiple-input multiple-output (MIMO) approach for multi-step-ahead time series prediction with a Gaussian process regression (GPR) model. We assess the forecasting performance of the GPR model with respect to several neural network architectures. The MIMO setting allows modelling the cross-correlations between all regions simultaneously. We find that the radial basis function (RBF) network outperforms the GPR model, especially for long-term forecast horizons. As the memory of the models increases, the forecasting performance of the GPR improves, suggesting the convenience of designing a model selection criteria in order to estimate the optimal number of lags used for concatenation.
    Keywords: Regional forecasting, tourism demand, multiple-input multiple-output (MIMO), Gaussian process regression, neural networks, machine learning. JEL classification:
    Date: 2017–01
  6. By: AUGUSTYNIAK, Maciej (Université de Montréal); BAUWENS, Luc (Université catholique de Louvain, CORE, Belgium); DUFAYS, Arnaud (Université Laval)
    Abstract: A new model – the high-dimensional Markov (HDM) model – is proposed for financial returns and their latent variances. It is also applicable to model directly realized variances. Volatility is modeled as a product of three components: a Markov chain driving volatility persistence, an independent discrete process capable of generating jumps in the volatility, and a predictable (data-driven) process capturing the leverage effect. The Markov chain and jump components allow volatility to switch abruptly between thousands of states. The transition probability matrix of the Markov chain is structured in such a way that the multiplicity of the second largest eigenvalue can be greater than one. This distinctive feature generates a high degree of volatility persistence. The statistical properties of the HDM model are derived and an economic interpretation is attached to each component. In-sample results on six financial time series highlight that the HDM model compares favorably to the main existing volatility processes. A forecasting experiment shows that the HDM model significantly outperforms its competitors when predicting volatility over time horizons longer than five days.
    Keywords: Volatility, Markov-switching, Persistence, Leverage effect
    JEL: C22 C51 C58
    Date: 2016–12–13
  7. By: Maria Putintseva (University of Zurich, Ecole Polytechnique Fédérale de Lausanne, and Swiss Finance Institute)
    Abstract: I propose a class of hybrid models to describe and predict the dynamics of a multivariate stationary random vector, e.g. a vector of stock returns. These models combine essential features of the multivariate mixture normal distribution and the conditional correlation models. I describe in detail the expectation-maximization algorithm, which makes the parameter estimation feasible and fast virtually for any random vector length. I fit the suggested models to five data sets, consisting of vectors of stock returns, with the maximal vector length of fifteen stocks. The predictive ability of this model class is compared to other widely used multivariate models, and it turns out that my models provide the best forecasts, both on average and for extreme negative returns. All necessary formulas to apply these models for important financial objectives are also provided.
    Keywords: Finite Mixtures, Dynamic Conditional Correlation, Forecasting, Multivariate Modelling, Predictive Ability
    JEL: C51 C53 G17
  8. By: Gianluca Cubadda (DEF and CEIS, University of Rome "Tor Vergata"); Barbara Guardabascio (ISTAT)
    Abstract: We examine the conditions under which each individual series that is generated by a vector autoregressive model can be represented as an autoregressive model that is augmented with the lags of few linear combinations of all the variables in the system. We call this modelling Multivariate Index-Augmented Autoregression (MIAAR). We show that the parameters of the MIAAR can be estimated by a switching algorithm that increases the Gaussian likelihood at each iteration. Since maximum likelihood estimation may perform poorly when the number of parameters gets larger, we propose a regularized version of our algorithm to handle a medium-large number of time series. We illustrate the usefulness of the MIAAR modelling both by empirical applications and simulations.
    Keywords: Multivariate index autoregressive models, reduced rank regression, dimension reduction, shrinkage estimation, macroeconomic forecasting
    JEL: C32
    Date: 2017–02–07

This nep-for issue is ©2017 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.