
Yuichi Goto (Kyushu University) “ANOVATS: A subsampling-based test for differences among short time series in marine studies”
Finance-Insurance
Time: 11.00 am
Date:11th of September 2025
Room 3001
Yuichi Goto (Kyushu University) “ANOVATS: A subsampling-based test for differences among short time series in marine studies”
Abstract :
Data used in marine ecosystem assessments often consist of short time series, making conventional methods that rely on large datasets potentially prone to drawing incorrect conclusions. Analysis of variance (ANOVA) methods for time series data is one example. These methods require the estimation of the reciprocal of the long-run variance (the spectral density at zero frequency), which typically demands long-term time series observations. In this talk, we address the problem of testing the equality of means across regions and propose a subsampling-based method for short time series data. The method also facilitates divisive hierarchical clustering by iteratively splitting regions based on statistical criteria. Its application to North Sea zooplankton biomass data reveals significant differences across diverse ecosystems, highlighting its utility in marine resource management.
Organizers: Christian FRANCQ