![]() | Only 14 pages are availabe for public view |
Abstract Time series analysis based on the assumption that the structure of the series can be described by linear dynamic models assumed to be Gaussian processes. The reasons for linearity assumption are that these models were convenient for analyzing the data and they often provide a fairly good approximation to real life time series data. However there are many occasions when data suggests that linear models are not appropriate. In those cases it is desirable to look at non-linear alternatives. One of the class of non-linear models known as bilinear models for which many analytical properties can be obtained and have received a great deal of interest. |