nep-ets New Economics Papers
on Econometric Time Series
Issue of 2021‒02‒15
twelve papers chosen by
Jaqueson K. Galimberti
Auckland University of Technology

  1. Peaks, Gaps, and Time Reversibility of Economic Time Series By Tommaso Proietti
  2. The accuracy of asymmetric GARCH model estimation By Olivier Darné; Amélie Charles
  3. A Test of Sufficient Condition for Infinite-step Granger Noncausality in Infinite Order Vector Autoregressive Process By Umberto Triacca; Olivier Damette; Alessandro Giovannelli
  4. Graphical Models for Financial Time Series and Portfolio Selection By Ni Zhan; Yijia Sun; Aman Jakhar; He Liu
  5. Unraveling S&P500 stock volatility and networks -- An encoding and decoding approach By Xiaodong Wang; Fushing Hsieh
  6. Forecasting Commodity Markets Volatility: HAR or Rough? By Mesias Alfeus; Christina Sklibosios Nikitopoulos
  7. On an integer-valued stochastic intensity model for time series of counts By Aknouche, Abdelhakim; Dimitrakopoulos, Stefanos
  8. The Economic Impact of Volatility Persistence on Energy Markets By Christina Sklibosios Nikitopoulos; Alice Thomas; Jianxin Wang
  9. Modelling and forecasting inflation rate in Nigeria using ARIMA models By Olalude, Gbenga Adelekan; Olayinka, Hammed Abiola; Ankeli, Uchechi Constance
  10. Nowcasting Monthly GDP with Big Data: a Model Averaging Approach By Tommaso Proietti; Alessandro Giovannelli
  11. Wind Generation and the Dynamics of Electricity Prices in Australia By Muthe Mathias Mwampashi; Christina Sklibosios Nikitopoulos; Otto Konstandatos; Alan Rai
  12. Nowcasting GDP and its Components in a Data-rich Environment: the Merits of the Indirect Approach By Alessandro Giovannelli; Tommaso Proietti; Ambra Citton; Ottavio Ricchi; Cristian Tegami; Cristina Tinti

  1. By: Tommaso Proietti (DEF and CEIS, Università di Roma "Tor Vergata")
    Abstract: Locating the running maxima and minima of a time series, and measuring the current deviation from them, generates processes that are analytically relevant for the analysis of the business cycle and for characterizing bull and bear phases in financial markets. The measurement of the time distance from the running peak originates a first order Markov chain, whose characteristics can be used for testing time reversibility of economic dynamics and specific types of asymmetries in financial markets. The paper derives the time series properties of the gap process and other related processes that arise from the same measurement context, and proposes new nonparametric tests of time reversibility. Empirical examples illustrate their uses for characterizing the depth of a recession and the duration of bull and a bear market.
    Keywords: Markov chains, Business cycles, Recession duration.
    JEL: C22 C58 E32
    Date: 2020–06–17
  2. By: Olivier Darné (LEMNA - Laboratoire d'économie et de management de Nantes Atlantique - IEMN-IAE Nantes - Institut d'Économie et de Management de Nantes - Institut d'Administration des Entreprises - Nantes - UN - Université de Nantes - IUML - FR 3473 Institut universitaire Mer et Littoral - UBS - Université de Bretagne Sud - UM - Le Mans Université - UA - Université d'Angers - CNRS - Centre National de la Recherche Scientifique - IFREMER - Institut Français de Recherche pour l'Exploitation de la Mer - UN - Université de Nantes - ECN - École Centrale de Nantes); Amélie Charles (Audencia Business School - Audencia Business School)
    Date: 2020–12–28
  3. By: Umberto Triacca (University of L'Aquila); Olivier Damette (University of Lorraine); Alessandro Giovannelli (University of L'Aquila)
    Abstract: This paper derives a sufficient condition for noncausality at all forecast horizons (infinitestep noncausality). We propose a test procedure for this sufficient condition. Our procedure presents two main advantages. First, our infinite-step Granger causality analysis is conducted in a more general framework with respect to the procedures proposed in literature. Second, it involves only linear restrictions under the null, that can be tested by using standard F statistics. A simulation study shows that the proposed procedure has reasonable size and good power. Typically, one thousand or more observations are required to ensure that the test procedures perform reasonably well. These are typical sample sizes for financial time series applications. Here, we give a first example of possible applications by considering the Mixture Distribution Hypothesis in the Foreign Exchange Market
    Keywords: Granger causality,Hypothesis testing,Time series,Vector autoregressive Models
    JEL: C12 C22 C58
    Date: 2020–06–18
  4. By: Ni Zhan; Yijia Sun; Aman Jakhar; He Liu
    Abstract: We examine a variety of graphical models to construct optimal portfolios. Graphical models such as PCA-KMeans, autoencoders, dynamic clustering, and structural learning can capture the time varying patterns in the covariance matrix and allow the creation of an optimal and robust portfolio. We compared the resulting portfolios from the different models with baseline methods. In many cases our graphical strategies generated steadily increasing returns with low risk and outgrew the S&P 500 index. This work suggests that graphical models can effectively learn the temporal dependencies in time series data and are proved useful in asset management.
    Date: 2021–01
  5. By: Xiaodong Wang; Fushing Hsieh
    Abstract: We extend the Hierarchical Factor Segmentation(HFS) algorithm for discovering multiple volatility states process hidden within each individual S&P500 stock's return time series. Then we develop an associative measure to link stocks into directed networks of various scales of associations. Such networks shed lights on which stocks would likely stimulate or even promote, if not cause, volatility on other linked stocks. Our computing endeavors starting from encoding events of large return on the original time axis to transform the original return time series into a recurrence-time process on discrete-time-axis. By adopting BIC and clustering analysis, we identify potential multiple volatility states, and then apply the extended HFS algorithm on the recurrence time series to discover its underlying volatility state process. Our decoding approach is found favorably compared with Viterbi's in experiments involving both light and heavy tail distributions. After recovering the volatility state process back to the original time-axis, we decode and represent stock dynamics of each stock. Our measurement of association is measured through overlapping concurrent volatility states upon a chosen window. Consequently, we establish data-driven associative networks for S&P500 stocks to discover their global dependency relational groupings with respect to various strengths of links.
    Date: 2021–01
  6. By: Mesias Alfeus; Christina Sklibosios Nikitopoulos (Finance Discipline Group, UTS Business School, University of Technology Sydney)
    Abstract: Commodity is one of the most volatile markets and forecasting its volatility is an issue of paramount importance. We study the dynamics of the commodity markets volatility by employing fractional stochastic volatility and heterogeneous autoregressive (HAR) models. Based on a high-frequency futures price dataset of 22 commodities, we confirm that the volatility of commodity markets is rough and volatility components over different horizons are economically and statistically significant. Long memory with anti-persistence is evident across all commodities, with weekly volatility dominating in most commodity markets and daily volatility for oil and gold markets. HAR models display a clear advantage in forecasting performance compared to fractional volatility models.
    Keywords: commodity markets; realized volatility; fractional Brownian motion; HAR; volatility forecast
    JEL: C20 C53 C58 G13 Q02
    Date: 2020–12–01
  7. By: Aknouche, Abdelhakim; Dimitrakopoulos, Stefanos
    Abstract: We propose a broad class of count time series models, the mixed Poisson integer-valued stochastic intensity models. The proposed specification encompasses a wide range of conditional distributions of counts. We study its probabilistic structure and design Markov chain Monte Carlo algorithms for two cases; the Poisson and the negative binomial distributions. The methodology is applied to simulated data as well as to various data sets. Model comparison using marginal likelihoods and forecast evaluation using point and density forecasts are also considered.
    Keywords: Markov chain Monte Carlo, mixed Poisson process, parameter-driven models, count time series models.
    JEL: C11 C13 C15 C18 C25 C5 C51 C53 C63
    Date: 2020–01–01
  8. By: Christina Sklibosios Nikitopoulos (Finance Discipline Group, UTS Business School, University of Technology Sydney); Alice Thomas (Finance Discipline Group, UTS Business School, University of Technology Sydney); Jianxin Wang (Finance Discipline Group, UTS Business School, University of Technology Sydney)
    Abstract: This study examines the role of daily volatility persistence in determining future volatility in energy markets. In crude oil and natural gas markets, the impact of returns and variances is primarily transmitted to future volatility via the daily volatility persistence. Macro-economic factors, such as the VIX, the credit spread and the Baltic exchange dirty index, also impact future volatility, but this impact is again channeled via the volatility persistence. The dependence of volatility persistence on macro-economic conditions is termed conditional volatility persistence (CVP). The variation in daily CVP is economically significant, contributing up to 17% of future volatility and accounting for 25% of the model's explanatory power. Inclusion of the CVP in the model significantly improves volatility forecasts. Based on the utility benefits of volatility forecasts, the CVP adjusted volatility models provide up to 160 bps benefit to investors compared to the HAR models, even after accounting for transaction costs and varying trading speeds.
    Keywords: Realized Volatility; Volatility Persistence; Energy Markets; HAR; Forecasting
    JEL: C22 C53 C58 Q40
    Date: 2020–12–01
  9. By: Olalude, Gbenga Adelekan; Olayinka, Hammed Abiola; Ankeli, Uchechi Constance
    Abstract: This study modelled and forecast inflation in Nigeria using the monthly Inflation rate series that spanned January 2003 to October 2020 and provided three years monthly forecast for the inflation rate in Nigeria. We examined 169 ARMA, 169 ARIMA, 1521 SARMA, and 1521 SARIMA models to identify the most appropriate model for modelling the inflation rate in Nigeria. Our findings indicate that out of the 3380 models examined, SARMA (3, 3) x (1, 2)12 is the best model for forecasting the monthly inflation rate in Nigeria. We selected the model based on the lowest Akaike Information Criteria (AIC) and Schwarz Information Criterion (SIC) values, volatility, goodness of fit, and forecast accuracy measures, such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The AIC and SIC of the model are 3.3992 and 3.5722, respectively with an adjusted R2 value of 0.916. Our diagnostic tests (Autocorrelation and Normality of Residuals) and forecast accuracy measures indicate that the presented model, SARMA (3, 3)(1, 2)12, is good and reliable for forecasting. Finally, the three years monthly forecast was made, which shows that the Inflation rate in Nigeria would continue to decrease but maintain a 2 digits value for the next two years, but is likely to rise again in 2023. This study is of great relevance to policymakers as it provides a foresight of the likely future inflation rates in Nigeria. Keywords: Inflation; Modelling, Forecasting; ARMA; ARIMA; SARMA; SARIMA;
    Keywords: Inflation; Modelling; Forecasting; ARMA; ARIMA; SARMA; SARIMA
    JEL: C22 C52 C53 E31 E37 E47
    Date: 2020
  10. By: Tommaso Proietti (CEIS & DEF, University of Rome "Tor Vergata"); Alessandro Giovannelli (DEF, University of Rome "Tor Vergata")
    Abstract: Gross domestic product (GDP) is the most comprehensive and authoritative measure of economic activity. The macroeconomic literature has focused on nowcasting and forecasting this measure at the monthly frequency, using related high frequency indicators. We address the issue of estimating monthly gross domestic product using a large dimensional set of monthly indicators, by pooling the disaggregate estimates arising from simple and feasible bivariate models that consider one indicator at a time in conjunction to GDP. Our base model handles mixed frequency data and ragged-edge data structure with any pattern of missingness. Our methodology enables to distill the common component of the available economic indicators, so that the monthly GDP estimates arise from the projection of the quarterly figures on the space spanned by the common component. The weights used for the combination reflect the ability to nowcast quarterly GDP and are obtained as a function of the regularized estimator of the high-dimensional covariance matrix of the nowcasting errors. A recursive nowcasting and forecasting experiment illustrates that the optimal weights adapt to the information set available in real time and vary according to the phase of the business cycle.
    Keywords: Mixed-Frequency Data, Dynamic Factor Models, State Space Models,Shrinkage
    JEL: C32 C52 C53 E37
    Date: 2020–05–12
  11. By: Muthe Mathias Mwampashi (Finance Discipline Group, UTS Business School, University of Technology Sydney); Christina Sklibosios Nikitopoulos (Finance Discipline Group, UTS Business School, University of Technology Sydney); Otto Konstandatos (Finance Discipline Group, UTS Business School, University of Technology Sydney); Alan Rai (University of Technology Sydney)
    Abstract: Australia's National Electricity Market (NEM) is experiencing one of the world's fastest and marked transitions toward variable renewable energy generation. This transformation poses challenges to system security and reliability and has triggered increased variability and uncertainty in electricity prices. By employing an exponential generalized autoregressive conditional heteroskedasticity (eGARCH) model, we gauge the effects of wind power generation on the dynamics of electricity prices in the NEM. We find that a 1 GWh increase in wind generation decreases daily prices up to 1.3 AUD/MWh and typically increases price volatility up to 2%. Beyond consumption and gas prices, hydro generation also contributes to an increase in electricity prices and their volatility. The cross-border interconnectors play a significant role in determining price levels and volatility dynamics. This underscores the important role of strategic provisions and investment in the connectivity within the NEM to ensure the reliable and effective delivery of renewable energy generation. Regulatory interventions, such as the carbon pricing mechanism and nationwide lockdown restrictions due to COVID-19 pandemic, also had a measurable impact on electricity price dynamics.
    Keywords: wind generation; electricity price volatility; merit order effect; hydro generation; interconnectors; carbon pricing mechanism; COVID-19
    JEL: C22 C58 Q40 Q42
    Date: 2020–12–01
  12. By: Alessandro Giovannelli (University of L'Aquila); Tommaso Proietti (DEF & CEIS, Università di Roma "Tor Vergata"); Ambra Citton (Ministero dell'Economia e delle Finanze); Ottavio Ricchi (Ministero dell'Economia e delle Finanze); Cristian Tegami (Sogei SpA); Cristina Tinti (Ministero dell'Economia e delle Finanze)
    Abstract: The national accounts provide a coherent and exaustive description of the current state of the economy, but are available at the quarterly frequency and are released with a nonignorable publication lag. The paper proposes and illustrates a method for nowcasting and forecasting the sixteen main components of Gross Domestic Product (GDP) by output and expenditure type at the monthly frequency, using a high-dimensional set of monthly economic indicators spanning the space of the common macroeconomic and financial factors. The projection on the common space is carried out by combining the individual nowcasts and forecasts arising from all possible bivariate models of the unobserved monthly GDP component and the observed monthly indicator. We discuss several pooling strategies and we select the one showing the best predictive performance according to a pseudo real time forecasting experiment. Monthly GDP can be indirectly estimated by the contemporaneous aggregation of the value added of the different industries and of the expenditure components. This enables the comparative assessment of the indirect nowcasts and forecasts vis-à-vis the direct approach and a growth accounting exercise. Our approach meets the challenges posed by the dimensionality, since it can handle a large number of time series with a complexity that increases linearly with the cross-sectional dimension, while retaining the essential heterogeneity of the information about the macroeconomy. The application to the Italian case leads to several interesting discoveries concerning the time-varying predictive content of the information carried by the monthly indicators.
    Keywords: Mixed-Frequency Data, Dynamic Factor Models, Growth Accounting, Model Averaging, Ledoit-Wolf Shrinkage.
    JEL: C32 C52 C53 E37
    Date: 2020–05–30

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.