Research
Realized GARCH: A Joint Model of Return and
Realized Measures of Volatility (with P. Hansen and H. Shek)
(Job Market Paper)
GARCH
models have been successful in modeling financial returns. Still, much
is to be gained by incorporating a realized measure of volatility in
these models. In this paper we introduce a new framework for the joint
modeling of returns and realized measures of volatility. The Realized
GARCH framework nests standard GARCH models as special cases and is, in
many ways, a natural extension of standard GARCH models. We pay special
attention to Realized GARCH models with linear and log-linear
specifications of the GARCH and measurement equations. This class of
models has several attractive features. It retains the simplicity and
tractability of the classical GARCH framework; it implies an ARMA
structure for the conditional variance and realized measures of
volatility; and models in this class are parsimonious and simple to
estimate. A key feature of the Realized GARCH framework is a
measurement equation that relates the observed realized measure to
latent volatility. This equation facilitates a simple modeling of the
dependence between returns and future volatility that is commonly
referred to as the leverage effect. We derive the asymptotic properties
of the QMLE estimator and show that it has a Gaussian limit
distribution. An empirical application with DJIA stocks and an exchange
traded index fund shows that a simple Realized GARCH structure leads to
substantial improvements in the empirical fit over to the standard
GARCH model. This is true in-sample as well as out-of-sample. Moreover,
the point estimates are remarkable similar across the different time
series.
The Impact of Market Structure Changes and Increasing Environmental Concerns on the Heat Arbitrage Relationship between Natural Gas and
Crude Oil Prices
(selected
as the Best PhD Candidacy Paper Award in the Economics Department at Stanford
in 2007)
Natural gas and crude oil are substitute
fuels in industrial, manufacturing and residential sectors, suggesting
a
possible heat arbitrage relationship between natural gas and oil
prices. However,
the extent to which natural gas can substitute for oil has been
historically
limited by the market regulation, additional capital costs and supply
security
of natural gas. This explains, in the US, why natural gas was
sold at a
lower price per heat unit than oil. Changes in market structures and
increasing
environmental concerns have favored the use of natural gas as fuel
since the
1990s and therefore changed the natural gas and oil prices dynamics. In
this
paper, I use cointegration models with structural change to identify
this effect.
The empirical results, consistent to my economic explanations, show that (i) there exists a long run cointegration
relationship between US oil and natural gas prices in which oil prices
lead gas
prices; (ii) during the 1991-2006 period, the relative prices of
natural gas to
oil have been significantly greater than those of 1976-1990; (iii)
natural gas
prices moved significantly closer to oil prices after 1990, owing to
the
increased fuel switching flexibility between natural gas and oil in the
power
generation and manufacturing sectors.
The Real-time Announcement Effects of Crude Oil Inventory Changes on Oil Futures Return and Volatility
(work in progress)
The working and
efficiency of commodity futures exchanges has been a focus of debate
during recent years. Using high frequency oil futures trading prices,
this paper documents three empirical facts: (i) real-time volatilities
for oil
futures increase before and after the weekly inventory announcements
and jumps
typically follow large inventory changes. (ii) Futures prices only
respond to
inventory change surprises but not expected changes measured by market
analyst
consensus. (iii) The futures market absorbs new information
very fast: the response vanishes within 10 minutes, even faster than the 15 minutes of stock market index.
|
