Skip to main content
Conference abstract, 2007

Subsampling Variance Estimation

Ekström, Magnus


Most proposed subsampling and resampling methods in the literature assume stationary data. In many empirical applications, however, the hypothesis of stationarity can easily be rejected. In this talk, it is demonstrated that variance estimators based on the subsampling methodology can be employed for different types of nonstationarity data, and the estimators are shown to be consistent under mild moment and mixing conditions. Results from a small simulation study on finite sample properties are provided, and an example with applications to forestry, using satellite data, is discussed

Published in


Exploring Stochastics