Statistics and finance : an interface : proceedings of the Hong Kong International Workshop on Statistics and Finance, Centre of Financial Time Series, the University of Hong Kong, 4-8 July 1999

This volume includes topics such as: invariants of strongly pseudoconvex CR manifolds; the integral formulas of the Pontrjagin characteristic forms on an oriented differentiable manifold; the construction of tensor fields and connections on the frame bundle; and cellular manufacturing systems Contents:Heavy-Tailed and Nonlinear Continuous-Time ARMA Models for Financial Time Series (P J Brockwell)Nonlinear State Space Model Approach to Financial Time Series with Time-Varying Variance (G Kitagawa & S Sato)Nonparametric Estimation and Bootstrap for Financial Time Series (J-P Kreiβ)A Note on Kernel Estimation in Integrated Time Series (Y-C Xia et al.)Stylized Facts on the Temporal and Distributional Properties of Absolute Returns: An Update (C W J Granger et al.)Volatility Computed by Time Series Operators at High Frequency (U A Müller)Missing Values in ARFIMA Models (W Palma)Second Order Tail Effects (C G de Vries)Bayesian Estimation of Stochastic Volatility Model via Scale Mixtures Distributions (S T B Choy & C M Chan)On a Smooth Transition Double Threshold Model (Y N Lee & W K Li)Interval Prediction of Financial Time Series (B Cheng & H Tong)A Decision Theoretic Approach to Forecast Evaluation (C W J Granger & M H Pesaran)Portfolio Management and Market Risk Quantification Using Neural Networks (J Franke)Detecting Structural Changes Using Genetic Programming with an Application to the Greater-China Stock Markets (X B Zhang et al.)and other papersReadership: Researchers in finance, time series analysis, economics and actuarial science, as well as investment bankers, stock market analysts and risk managers. This volume constitutes the proceedings of the Hong Kong International Workshop on Statistics in Finance, held at the University of Hong Kong in July 1999. Topics covered include heavy-tailed and nonlinear continuous-time ARMA models for financial time series, and forecast evaluation. Properties of linear continuous-time ARMA (or CARMA) processes driven by second-order Levy processes are examined.