nach oben

Empirical Economics

Erschienen in:

17.10.2018

Realized volatility and jump testing in the Japanese electricity spot market

verfasst von: Aitor Ciarreta, Peru Muniain, Ainhoa Zarraga

Erschienen in: Empirical Economics | Ausgabe 3/2020

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

The analysis of price volatility in electricity markets is increasingly significant for market participants. Realized measures have proved to be a useful tool. In this paper, we analyze realized volatility calculated from half-hour electricity prices on the Japanese spot market for the period from April 2005 to December 2015. Our interest stems from the fact that Japan is an isolated country with an electricity market in the process of being deregulated. We apply six alternative jump tests available in the literature to decompose total realized variation into jump and continuous components. We find large differences from one test to another in the number of jump-days identified arising from the nature of the data and the characteristics of the tests. We then estimate several heterogeneous autoregressive models for total and decomposed realized volatility and also consider GARCH innovations. Our results show high persistence of volatility and significant jumps. Finally, we assess the performance and forecasting ability of the models using in-sample and out-of-sample criteria. The model selected with both types of criteria includes the jumps obtained using the Jiang and Oomen (J Econom 144:352–370, 2008) jump test as regressors together with lagged total variation and GARCH innovations. Our results are significant in helping participants in the Japanese electricity market to take optimal decisions based on price characteristics.

Vorheriger Artikel On the relationship between political alignment and government transfers: triple differences evidence from a developing country

Nächster Artikel Forecasting precious metal returns with multivariate random forests

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

In cost-of-service regulation, the regulator determines the total revenue that must be collected in rates for the utility to recover the cost of providing the service and earn a reasonable return on investment. There is no price signal for scarcity of resources.

The price in the pool is determined by matching selling offers from suppliers and purchase bids from buyers at time intervals. In Australia and Japan, these prices are set for each half-hour and for France, Germany, Italy and Spain they are set for each hour.

There are other jump tests such as Aït-Sahalia and Jacod (2009), Andersen et al. (2007a), Andersen et al. (2010) and Mancini (2009). However, some of them are based on measures of IV similar to those applied in the paper or are generally used for detecting intraday jumps, as in the case of the LM test.

Under yardstick regulation, the performance of a regulated utility is compared to a group of comparable utilities (Shleifer 1985) Thus, the price cap of each utility is determined in a way that can end up in losses. This system is expected to promote efficiency through cost cutting.

The exchange operator does not open real price trading. It opens a system marginal price for the operating regions and a virtual one for the whole country (assuming no restrictions in the network and no differences in the frequencies).

Since then, most EPCOs have been running at a loss.

In order to remove seasonality, two different band-pass filters are also applied: Baxter and King (1999) and Christiano and Fitzgerald (2003). Results are similar to those using the simple de-median approach. This is why we decide to stick to the simplest approach.

Tauchen and Zhou (2011) show that \(\alpha \) should be chosen according to the number of jumps. The higher the contribution of jumps to the total variation, the smaller the \(\alpha \) chosen should be.

Extreme supply shocks such as the Fukushima–Daiichi nuclear disaster did not have an special impact on returns because the shock was not passed on into large price fluctuations. This is why tests do not detect any jump in that particular day.

RV is a highly autocorrelated series with an autocorrelation function that decays at a very slow rate. Lagged RV, JV and CV for horizons longer than a month are also included in the models as explanatory variables, but as they are not significant the maximum horizon considered is 1 month. Results are available upon request.

Results are available upon request.

The estimation results for the rest of the GARCH-type innovations are not reported since the estimated coefficients of the GARCH equation do not satisfy the conditions for a finite unconditional variance and models with IGARCH innovations are always preferred to those with EGARCH innovations. The only exception is when the MIN test is used since the HAR–CV–JV model with EGARCH innovations is preferred to that with IGARCH innovations. However, the former is not selected among the other models. Results are available from the authors upon request.

\(\mathrm{RMSE} = \sqrt{\frac{1}{F}\sum _{t=1}^F (\mathrm{RV}_t - \widehat{\mathrm{RV}_t})^2}\), where F is the number of forecast observations and \(\widehat{\mathrm{RV}_t}\) is the predicted value of RV. MAE (mean absolute error), MAPE (mean absolute percentage error) criteria are also calculated, and the lowest value is obtained for the same model as when using RMSE criterion. Results are available upon request.

We also calculate the RMSE criterion from a recursive estimation of the models using the first observations and adding one new observation in each step. The selection of the models does not change. Results are available upon request.

Results are available from the authors upon request.

Aït-Sahalia Y, Jacod J (2009) Testing for jumps in a discretely observed process. Ann Stat 37(1):184–222CrossRef

Andersen TG, Bollerslev T (1998) Answering the skeptics: yes, standard volatility models do provide accurate forecasts. Int Econ Rev 39:885–905CrossRef

Andersen TG, Bollerslev T, Diebold FX (2007a) Roughing it up: including jump components in the measurement, modeling, and forecasting of return volatility. Rev Econ Stat 89(4):701–720CrossRef

Andersen TG, Bollerslev T, Diebold FX, Labys P (2003) Modelling and forecasting realized volatility. Econometrica 71:579–625CrossRef

Andersen TG, Bollerslev T, Dobrev D (2007b) No-arbitrage semi-martingale restrictions for continuous-time volatility models subject to leverage effects, jumps and i.i.d. noise: Theory and testable distributional implications. J Econom 138:125–180CrossRef

Andersen TG, Bollerslev T, Frederiksen P, Nielsen MØ (2010) Continuous-time models, realized volatilities, and testable distributional implications for daily stock returns. J Appl Econom 25:233–261CrossRef

Andersen TG, Dobrev D, Schaumburg E (2012) Jump-robust volatility estimation using nearest neighbor truncation. J Econom 169:75–93CrossRef

Barndorff-Nielsen O, Shephard N (2004) Power and bipower variation with stochastic volatility and jumps. J Financ Econom 2(1):1–37CrossRef

Barndorff-Nielsen O, Shephard N (2006) Econometrics of testing for jumps in financial economics using bipower variation. J Financ Econom 4(1):1–30CrossRef

Baxter M, King RG (1999) Measuring business cicles: approximate band-pass filters for economic time series. Rev Econ Stat 81(4):575–593CrossRef

Chan KF, Gray P, Van Campen B (2008) A new approach to characterizing and forecasting electricity volatility. Int J Forecast 24:728–743CrossRef

Christiano LJ, Fitzgerald TJ (2003) The band pass filter. Int Econ Rev 44(2):435–465CrossRef

Ciarreta A, Zarraga A (2016) Modeling realized volatility on the Spanish intra-day electricity market. Energy Econ 58:152–163CrossRef

Ciarreta A, Muniain P, Zarraga A (2017) Modeling and forecasting realized volatility in German–Austrian continuous intraday electricity prices. J Forecast 36:680–690CrossRef

Corsi F (2009) A simple approximate long-memory model of realized volatility. J Financ Econom 7(2):174–196CrossRef

Corsi F, Mittnik S, Pigorsch C, Pigorsch U (2008) The volatility of realized volatility. Econom Rev 27:46–78CrossRef

Corsi F, Pirino D, Reno R (2010) Threshold bipower variation and the impact of jumps on volatility forecasting. J Econom 159:276–288CrossRef

Cramton P (2017) Electricity market design. Oxf Rev Econ Policy 33(4):589–612CrossRef

Dumitru A-M, Urga G (2012) Identifying jumps in financial assets: a comparison between nonparametric jump tests. J Bus Econ Stat 30(2):242–255CrossRef

Haugom E, Westgaard S, Solibakke PB, Lien G (2011) Realized volatility and the influence of market measures on predictability: analysis of Nord Pool forward electricity data. Energy Econ 33(6):1206–1215CrossRef

Hausman JA (1978) Specification tests in econometrics. Econometrica 46:1251–1271CrossRef

Huang X, Tauchen G (2005) The relative contribution of jumps to total price variance. J Financ Econom 3(4):456–499CrossRef

Japanese Electricity Power Exchange. http://www.jepx.org/english/. Accessed 10 July 2016

Jiang GJ, Oomen RCA (2008) Testing for jumps when asset prices are observed with noise: a “swap varian” approach. J Econom 144:352–370CrossRef

Lee SS, Mykland PA (2008) Jumps in financial markets: a new nonparametric test and jump dynamics. Rev Financ Stud 21(6):2535–2563CrossRef

Mancini C (2009) Non-parametric threshold estimation for models with stochastic diffusion coefficient and jumps. Scand J Stat 36:270–296CrossRef

Ofuji K, Tatsumi N (2016) A count model analysis of traded contracts in the JEPX forward market. Electr Eng Jpn 194(4):10–18CrossRef

Qu H, Chen W, Niua M, Li X (2016) Forecasting realized volatility in electricity markets using logistic smooth transition heterogeneous autoregressive models. Energy Econ 54:68–76CrossRef

Shleifer A (1985) A theory of Yardstick competition. Rand J Econ 16(3):319–327CrossRef

Tauchen G, Zhou H (2011) Realized jumps on financial markets and predicting credit spreads. J Econom 160(1):102–118CrossRef

Ullrich CJ (2012) Realized volatility and price spikes in electricity markets: the importance of observation frequency. Energy Econ 34:1809–1818CrossRef

Titel: Realized volatility and jump testing in the Japanese electricity spot market
verfasst von: Aitor Ciarreta
Peru Muniain
Ainhoa Zarraga
Publikationsdatum: 17.10.2018
Verlag: Springer Berlin Heidelberg
Erschienen in: Empirical Economics / Ausgabe 3/2020
Print ISSN: 0377-7332
Elektronische ISSN: 1435-8921
DOI: https://doi.org/10.1007/s00181-018-1577-6

Springer Professional

Abstract

Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Springer Professional "Wirtschaft+Technik"

Weitere Artikel der Ausgabe 3/2020

Hysteresis losses in the Preisach framework

Drink and smoke; drink or smoke? The interdependence between alcohol and cigarette consumption for men in China

Social capital: exploring the theory and empirical divide

On the distribution of links in financial networks: structural heterogeneity and functional

Network analysis to measure academic performance in economics

Forecasting precious metal returns with multivariate random forests

Premium Partner