In this paper we have defined a new class of non-linear lime series models, e.g., threshold moving average model. These models are helpful in forecasting economic, financial and environmental time series data when the linear model fails. We have obtained the moments and distribution of threshold models. These results are found to be helpful in the specification of these models.
|Number of pages||8|
|Journal||Far East Journal of Mathematical Sciences|
|Publication status||Published - 1996|