nep-ets New Economics Papers
on Econometric Time Series
Issue of 2021‒04‒19
ten papers chosen by
Jaqueson K. Galimberti
Auckland University of Technology

  1. Application of Bagging in Day-Ahead Electricity Price Forecasting and Factor Augmentation By Kadir Özen; Dilem Yıldırım
  2. Forecasting UK inflation bottom up By Joseph, Andreas; Kalamara, Eleni; Kapetanios, George; Potjagailo, Galina
  3. Score-driven time series models By Harvey, A.
  4. Time Series (re)sampling using Generative Adversarial Networks By Christian M. Dahl; Emil N. S{\o}rensen
  5. Improved Tests for Granger Non-Causality in Panel Data By Xiao, Jiaqi; Juodis, Arturas; Karavias, Yiannis; Sarafidis, Vasilis
  6. Factor Models with Local Factors—Determining the Number of Relevant Factors By Simon Freyaldenhoven
  7. A newmacro-financial condition index for the euro area By Claudio Morana
  8. Improving the Estimation and Predictions of Small Time Series Models By Gareth Liu-Evans
  9. Efficient and robust inference of models with occasionally binding constraints By Giovannini, Massimo; Pfeiffer, Philipp; Ratto, Marco
  10. Recurrent Dictionary Learning for State-Space Models with an Application in Stock Forecasting By Shalini Sharma; Víctor Elvira; Emilie Chouzenoux; Angshul Majumdar

  1. By: Kadir Özen (Barcelona Graduate School of Economics, Barcelona, Spain); Dilem Yıldırım (Department of Economics, Middle East Technical University, Ankara, Turkey)
    Abstract: The electricity price forecasting (EPF) is a challenging task not only because of the uncommon characteristics of electricity but also because of the existence of many potential predictors with changing predictive abilities over time. Particularly, how to account for all available factors and extract as much information as possible is the key to the production of accurate forecasts. To address this long-standing issue in a way that balances complexity and forecasting accuracy while facilitating the traceability of the predictor selection procedure, the method of Bootstrap Aggregation (bagging), which is a variant shrinkage estimation approach for the estimation of large scale models, is proposed in this paper. To forecast day-ahead electricity prices in a multivariate context for six major power markets we construct a large scale pure-price model (in addition to some stochastic models that are commonly applied in the literature) and apply the bagging approach in comparison with the popular Least Absolute Shrinkage and Selection Operator (LASSO) estimation method. Our forecasting study reveals that with its superior forecasting performance and its computationally simple algorithm, the bagging emerges as a strong competitor to the commonly applied LASSO approach for the short-term EPF. Further analysis for the variable selection for the bagging and LASSO approaches suggests that the differentiation in the forecast performances of two approaches might be due to, inter alia, their structural differences in the explanatory variables selection process. Moreover, to account for the intraday hourly dependencies of day-ahead electricity prices, all our models are augmented with latent factors, and a substantial improvement is observed only in the forecasts from models covering a relatively limited number of predictors, while almost no improvement is obtained in the forecasts from the large scale model estimated through LASSO and bagging techniques.
    Keywords: Bagging, Shrinkage methods, Electricity price forecasting, Multivariate modeling, Forecast encompassing, Factor models
    JEL: C22 C38 C51 C53 Q47
    Date: 2021–04
  2. By: Joseph, Andreas (Bank of England); Kalamara, Eleni (King’s College London); Kapetanios, George (King’s College London); Potjagailo, Galina (Bank of England)
    Abstract: We forecast CPI inflation in the United Kingdom up to one year ahead using a large set of monthly disaggregated CPI item series combined with a wide set of forecasting tools, including dimensionality reduction techniques, shrinkage methods and non-linear machine learning models. We find that exploiting CPI item series over the period 2011–19 yields strong improvements in forecasting UK inflation against an autoregressive benchmark, above and beyond the gains from macroeconomic predictors. Ridge regression and other shrinkage methods perform best across specifications that include item-level data, yielding gains in relative forecast accuracy of up to 70% at the one-year horizon. Our results suggests that the combination of a large and relevant information set combined with efficient penalisation is key for good forecasting performance for this problem. We also provide a model-agnostic approach to address the general problem of model interpretability in high-dimensional settings based on model Shapley values, partial re-aggregation and statistical testing. This allows us to identify CPI divisions that consistently drive aggregate inflation forecasts across models and specifications, as well as to assess model differences going beyond forecast accuracy.
    Keywords: Inflation; forecasting; machine learning; state space models; CPI disaggregated data; Shapley values
    JEL: C32 C45 C53 C55 E37
    Date: 2021–03–26
  3. By: Harvey, A.
    Abstract: The construction of score-driven filters for nonlinear time series models is described and it is shown how they apply over a wide range of disciplines. Their theoretical and practical advantages over other methods are highlighted. Topics covered include robust time series modeling, conditional heteroscedasticity, count data, dynamic correlation and association, censoring, circular data and switching regimes.
    Keywords: copula, count data, directional data, generalized autoregressive conditional heteroscedasticity, generalized beta distribution of the second kind, observation-driven model, robustness
    JEL: C22 C32
    Date: 2021–04–07
  4. By: Christian M. Dahl; Emil N. S{\o}rensen
    Abstract: We propose a novel bootstrap procedure for dependent data based on Generative Adversarial networks (GANs). We show that the dynamics of common stationary time series processes can be learned by GANs and demonstrate that GANs trained on a single sample path can be used to generate additional samples from the process. We find that temporal convolutional neural networks provide a suitable design for the generator and discriminator, and that convincing samples can be generated on the basis of a vector of iid normal noise. We demonstrate the finite sample properties of GAN sampling and the suggested bootstrap using simulations where we compare the performance to circular block bootstrapping in the case of resampling an AR(1) time series processes. We find that resampling using the GAN can outperform circular block bootstrapping in terms of empirical coverage.
    Date: 2021–01
  5. By: Xiao, Jiaqi; Juodis, Arturas; Karavias, Yiannis; Sarafidis, Vasilis
    Abstract: This article introduces the xtgranger command in Stata, which implements the panel Granger non-causality test approach developed by Juodis, Karavias and Sarafidis (2021). This test offers superior size and power performance to existing tests, which stems from the use of a pooled estimator that has a faster √NT convergence rate. The test has two other useful properties; it can be used in multivariate systems and it has power against both homogeneous as well as heterogeneous alternatives.
    Keywords: Panel data, Granger non-causality, Nickell bias, Heterogeneous panels, Fixed effects, Half-panel Jackknife, xtgranger.
    JEL: C12 C23 C33
    Date: 2021–04–14
  6. By: Simon Freyaldenhoven
    Abstract: We extend the theory on factor models by incorporating “local” factors into the model. Local factors affect only an unknown subset of the observed variables. This implies a continuum of eigenvalues of the covariance matrix, as is commonly observed in applications. We de-rive which factors are pervasive enough to be economically important and which factors are pervasive enough to be estimable using the common principal component estimator. We then introduce a new class of estimators to determine the number of those relevant factors. Un-like existing estimators, our estimators use not only the eigenvalues of the covariance matrix, but also its eigenvectors. We find that incorporating partial sums of the eigenvectors into our estimators leads to significant gains in performance in simulations.
    Keywords: high-dimensional data; factor models; weak factors; local factors; sparsity
    JEL: C38 C52 C55
    Date: 2021–04–15
  7. By: Claudio Morana
    Abstract: In this paper, we introduce a new time-domain decomposition for weakly stationary or trend stationary processes, based on trigonometric polynomial modelling of the underlying component of an economic time series. The method is explicitly devised to disentangle medium to long-term and short-term fluctuations in macroeconomic and financial series, in order to accurately measure the financial cycle and the concurrent long swings in economic activity. The implementation of this decomposition is straightforward and relies on standard regression analysis and general to specific model reduction. Full support to the proposed method is provided by Monte Carlo simulation. In the paper, we also provide a multivariate extension, involving sequential univariate decompositions and Principal Components Analysis. Based on this multivariate approach, we introduce a set of new composite indexes of macro-financial conditions for the euro area and assess their information content. In particular, concerning the current pandemic, the indicators suggest that most of the GDP contraction has been of short-term, cyclical nature. This is likely due to the prompt monetary and fiscal policy responses. Yet our evidence suggests that the financial cycle might have currently achieved a peak area. Hence, the risk of further, deeper disruptions is high, particularly in so far as a new sovereign/corporate debt crisis were not eventually avoided.
    Keywords: trend-cycle decomposition, COVID-19 pandemics, subprime financial crisis, sovereign debt crisis, dot-com bubble, macroeconomic and financial conditions index, euro area
    JEL: C22 C38 E32 F44 G01
    Date: 2021–04
  8. By: Gareth Liu-Evans
    Abstract: A new approach is developed for improving the point estimation and predictions of para-metric time-series models. The method targets performance criteria such as estimation bias, root mean squared error, variance, or prediction error, and produces closed-form es-timators focused towards these targets via a computational approximation method. This is done for an autoregression coefficient, for the mean reversion parameter in Vasicek and CIR diffusion models, for the Binomial thinning parameter in integer-valued autoregres-sive (INAR) models, and for predictions from a CIR model. The success of the prediction targeting approach is shown in Monte Carlo simulations and in out-of-sample forecasting of the US Federal Funds rate.
    Date: 2021–01
  9. By: Giovannini, Massimo (European Commission); Pfeiffer, Philipp (European Commission); Ratto, Marco (European Commission)
    Abstract: This paper proposes a piecewise-linear Kalman filter (PKF) to estimate DSGE models with occasionally binding constraints. This method expands the set of models suitable for nonlinear estimation. It straightforwardly handles missing data, non-singularity (more shocks than observed time series), and large-scale models. We provide several applications to highlight its efficiency and robustness compared to existing methods. Our toolkit integrates the PKF into Dynare, the most popular software in DSGE modeling.
    Keywords: DSGE, occasionally binding constraints, nonlinear estimation, Piecewise Kalman Filter
    JEL: C11 C32 C51
    Date: 2021–04
  10. By: Shalini Sharma (IIIT-Delhi - Indraprastha Institute of Information Technology [New Delhi]); Víctor Elvira (School of Mathematics - University of Edinburgh - University of Edinburgh); Emilie Chouzenoux (OPIS - OPtimisation Imagerie et Santé - CVN - Centre de vision numérique - CentraleSupélec - Université Paris-Saclay - Inria - Institut National de Recherche en Informatique et en Automatique - Inria Saclay - Ile de France - Inria - Institut National de Recherche en Informatique et en Automatique); Angshul Majumdar (IIIT-Delhi - Indraprastha Institute of Information Technology [New Delhi])
    Abstract: In this work, we introduce a new modeling and inferential tool for dynamical processing of time series. The approach is called recurrent dictionary learning (RDL). The proposed model reads as a linear Gaussian Markovian state-space model involving two linear operators, the state evolution and the observation matrices, that we assumed to be unknown. These two unknown operators (that can be seen interpreted as dictionaries) and the sequence of hidden states are jointly learnt via an expectation-maximization algorithm. The RDL model gathers several advantages, namely online processing, probabilistic inference, and a high model expressiveness which is usually typical of neural networks. RDL is particularly well suited for stock forecasting. Its performance is illustrated on two problems: next day forecasting (regression problem) and next day trading (classification problem), given past stock market observations. Experimental results show that our proposed method excels over state-of-the-art stock analysis models such as CNN-TA, MFNN, and LSTM.
    Keywords: Stock Forecasting,Recurrent dictionary learning,Kalman filter,expectation-minimization,dynamical modeling,uncertainty quantification
    Date: 2021

This nep-ets issue is ©2021 by Jaqueson K. Galimberti. 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.