統計学輪講(第27回)

統計学輪講(第27回)
日時      2013年01月29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    朴 海香 (経済M1)
演題      Cholesky Stochastic Volatility Models for High-Dimensional Times
Series..

概要
Multivariate time-varying bolatility has many important applications in
finance.
 Estimating multivariate volatility, however, is not easy because of two
major difficulties.
 The first difficulty is the curse of dimensionality. The second difficulty
is
 that the conditional covariance matrix must be positive definite for all
time points.
 In ordet to simply maintain positive definiteness,
 Cholesky root of the time-varying covariance matrix model was proposed
by  H.F.Lopes, R.E.McCulloch and R.S.Tsay(2012).

We will mainly talk about Cholesky Stochastic Volatility Models
for High-Dimensional Times Series. We will also discuss the model's
sensitiveness
to the prior distribution and talk about the future works regarding this
problem.