Nehodí se? Vůbec nevadí! U nás můžete do 30 dní vrátit
S dárkovým poukazem nešlápnete vedle. Obdarovaný si za dárkový poukaz může vybrat cokoliv z naší nabídky.
30 dní na vrácení zboží
A new test to detect changes in the covariance §structure of a time series is developed. The test §does not involve direct fitting of an assumed model §for the time series. It is based on detecting §changes in autocovariances calculated in a moving §window through the series. The use of standard tests §of time series change points is inappropriate §because of the correlations imposed by the moving §windows. This requires the development of new §adjustments to existing time series change point §tests. The ability of this moving window technique §to detect changes in the lag one autocovariance of §autoregressive and moving average time series is §studied. We illustrate the application of this new §test on UK Treasury bill rates and airline travel §data.