×

Some applications of time series models to financial data. (English) Zbl 1228.62115

Summary: Data sets in economics and finance have often the form of time series. The article is devoted to an application of simple univariate and multivariate autoregressive models to a two-dimensional collection of exchange rates. Parameter estimates obtained using special methods constructed for non-negative time series are compared with the outputs of standard estimation procedures implemented in commonly used software products. Later on, the attention is paid to the predictive capability of our models.

MSC:

62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62P05 Applications of statistics to actuarial sciences and financial mathematics
91G70 Statistical methods; risk measures
62H12 Estimation in multivariate analysis
PDFBibTeX XMLCite