Modelling Financial Time Series
The most accurate and detailed time series models ever published, describing the behavior over time of stock, commodity and currency prices. Forty time series are investigated, including prices for stocks in New York and London, agricultural futures in Chicago, London, and Sydney, spot bullion and metal contracts in London and currency futures in Chicago. These prices are used to construct statistical models and to explore the benefits from relevant forecasts. Uses comprehensive and new models for price behavior.
What people are saying - Write a review
We haven't found any reviews in the usual places.
FEATURES OF FINANCIAL RETURNS
MODELLING PRICE VOLATILITY
FORECASTING STANDARD DEVIATIONS
8 other sections not shown
absolute returns approximately ARMACH assumed assumption asymptotic autocorrelation coefficients autocorrelations pT average buy and hold calculated Chapter Cocoa Cocoa 12 commodity and currency conditional standard deviation conditional variance consider constant correlation currency series daily returns defined denoted depend efficient market hypothesis equation estimates expected return Figure financial time series first-lag Firstly formula function futures contract futures series Gaussian given identical distributions investors kurtosis lags mean mean square errors method non-linear non-stationary normal distributions observations obtained optimal linear forecast option positive Praetz price-trend models product process random variables random walk hypothesis random walk tests reject rescaled returns returns process risk RMSE sample autocorrelations Section shows simulated spot series squared returns stationary models stationary process stochastic process stochastically independent strategy strict white noise suppose Swiss franc Table test statistics theoretical tions trading days trading rules trend models values variance changes X,+T zero