統計学輪講（第２５回）

統計学輪講（第25回）
日時      2013年01月08日(火)    15時40分～16時30分
場所      経済学部新棟３階第３教室
講演者    三崎 広海 (経済D3)
演題      An empirical analysis of volatility, covariance and hedging ratio
             by the SIML estimation at the Osaka Securities Exchange

概要
Recently a considerable interest has been paid on the estimation problem of
the integrated volatility by using high-frequency data in financial econometrics.
It has been well known that the conventional methods such as the realized volatility
and realized covariance work poorly when there exist market microstructure noise.

Kunitomo and Sato (2008a, b) have proposed the Separating Information Maximum
Likelihood (SIML) method for estimating the integrated volatility and variance under
the presence of market microstructure noise. The SIML method has been originally
defined on equidistant observations, but in actual markets the transactions occur randomly.

The main purpose of this report is to investigate the SIML estimation by using irregular
and non-synchronous high-frequency data. First we show that the SIML estimator has
reasonable robust properties in finite samples by conducting a number of Monte Carlo
simulations. Then we apply the SIML estimation to the transaction prices of individual
stocks traded at the Osaka Securities Exchange (OSE). We also estimate the hedging
ratio of the individual stocks by the Nikkei-225 Futures.