Multivariate Tests for Time Series Models
In their earlier, companion volume, Univariate Tests for Time Series Models (No. 99 of this series), the authors laid the groundwork for the explanation of tests for univariate stochastic (random) time series models. With such analysis, an essential first question is, "Does the stochastic proce...
Сохранить в:
| Главный автор: | |
|---|---|
| Формат: | |
| Язык: | English |
| Опубликовано: |
Sage Publications
2012
|
| Предметы: | |
| Online-ссылка: | https://scholar.dlu.edu.vn/thuvienso/handle/DLU123456789/30566 |
| Метки: |
Добавить метку
Нет меток, Требуется 1-ая метка записи!
|
| Thư viện lưu trữ: | Thư viện Trường Đại học Đà Lạt |
|---|
| Итог: | In their earlier, companion volume, Univariate Tests for Time Series Models (No. 99 of this series), the authors laid the groundwork for the explanation of tests for univariate stochastic (random) time series models. With such analysis, an essential first question is, "Does the stochastic process change over time?" If, for variable X, the answer is "yes," then it is nonstationary. A typical example of a nonstationary time series is United States Gross National Product (GNP), which has generally drifted upward since the 1930s (considering its more or less steadily rising mean value over the period). Upon reflection, it is obvious that many social science variables measured across time are nonstationary. Lamentably, uncritical analysis of nonstationary time series can generate spurious results. |
|---|