VOL. 10, NO 19, OCTOBER, 2015 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2015 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
8648
INDIAN ELECTRICITY MARKET VOLUME AND PRICE CROSS-
CORRELATION ANALYSIS
Mayukha Pal
1, 2, 3
, P. Madhusudana Rao
2
and P. Manimaran
1
1
C R Rao Advanced Institute of Mathematics, Statistics and Computer Science, University of Hyderabad Campus, Prof. C R Rao Road,
Hyderabad, India
2
College of Engineering, Jawaharlal Nehru Technological University, Hyderabad, India
3
India Innovation Center, General Electric Company, Secunderabad, India
E-Mail: maran@crraoaimscs.res.in
ABSTRACT
We apply the multifractal detrended cross-correlation analysis method to investigate the cross-correlation and
fractal behavior between the price and volume of the electricity market. For this purpose, we have collected the data from
the Indian electricity energy exchange from 1
st
April 2012 to 1
st
April 2015 with time interval of 1 hour. From the analysis,
we observe a cross over near the scale (~ 32) in the fluctuation function, and thus we have calculated the Hurst scaling
exponents for the scale <= 32 (short term) and > 32 (long term). The cross-correlation is observed persistent in short term
and anti-persistent in long term. The multifractal nature is present in both short and long term and the strength of the
multifractality was measured from the calculated singularity spectrum.
Keywords: electricity market, non-stationary time series, hurst exponent, multifractal detrended cross-correlation analysis, energy,
power data.
INTRODUCTION
Generation of sufficient electricity to cater
demand of the market is essential to the growth of any
developing nation. Availability of different source of
electricity power generation and grid efficiency plays an
important part of demand and supply chain that fuels
growth to the industry, employment, and infrastructure.
Effective, efficient supply chain and utilization of various
natural resources like hydro, thermal (coal, gas, diesel etc),
nuclear, renewable (wind, solar etc) etc. plays an
important part in electricity generation (Wang et al.,
2011), (Ramirez et al., 2009), (Uritskaya et al., 2008),
(Uritskaya et al., 2015).
For smooth supply chain and to provide
competitive business environment, electricity power
industry across the world is deregulated and liberalized.
Due to deregulation in electricity markets diverse
participants are involved like producers, intermediates,
speculators, traders, consumers hence the electricity
market price and traded volume is the result of many
interacting factors combined within complex and
nonlinear forms. Unlike any other traded commodities,
electricity power cannot be stored hence any un-utilized
demand is lost which is the unique feature in this market
that drives extreme volatility and large stochastic
dynamics into the future electricity price hence forecasting
becomes very difficult (Ramirez et al., 2010), (Malo et al.,
2009), (Wang et al., 2013), (Norouzzadeh et al., 2007),
(McArthur et al., 2013). Also additionally peak demand
and supply shortages drive extremely high price volatility
due to the complex, non-linear wholesale electricity
market dynamics. Hence with all these factor combination,
the electricity volume and price market becomes complex.
Thus analysis of electricity volume and price data
becomes very crucial to understand and retrieve critical
information from its underlying dynamics for its exhibited
behavior.
Accurate forecast of demand in electricity volume
and price would bring lot of stabilization in the electricity
market with highly improved supply chain vs. the demand
to cater the growing market. Hence any short-term volume
demand and price forecasts would help the producers
effectively plan the production, business efficiently (Fan et
al., 2015), (Wang et al., 2012), (Erzgraeber et al., 2008).
The forecast study and any future analysis are possible
only if the market behavior is well analyzed.
Fractal and correlation study has been an efficient
tool in analyzing the behavior of complex system. Study
of such complex dynamical system through the time series
data of the traded market using nonlinear dynamics
models are essential to understand the market behavior
(Peng et al., 1994), (Kantelhardt et al., 2002). Many
studies had undertaken in understanding the electricity
power market to study the correlation behavior and
multifractal nature of the real world recordings which are
in the form of time series so their multifractal properties
are extensively investigated (Manimaran et.al., 2005 and
refences there in). Similarly, the cross correlation behavior
of two non-stationary time series signals were investigated
using multifractal detrended cross correlation analysis
(MF-X-DFA) approach (Podobnik et.al., 2008), (Zhou,
2008), (Jiang et al., 2011), (Feng et al., 2013), (Pal et al.,
2014).