Sornborger Andrew T, Lauderdale James D
Department of Mathematics, University of California, Davis, CA.
Department of Cellular Biology, University of Georgia, Athens, GA.
Conf Rec Asilomar Conf Signals Syst Comput. 2016 Nov;2016:1056-1060. doi: 10.1109/ACSSC.2016.7869531. Epub 2017 Mar 6.
Neural data analysis has increasingly incorporated causal information to study circuit connectivity. Dimensional reduction forms the basis of most analyses of large multivariate time series. Here, we present a new, multitaper-based decomposition for stochastic, multivariate time series that acts on the covariance of the time series at all lags, (), as opposed to standard methods that decompose the time series, (), using only information at zero-lag. In both simulated and neural imaging examples, we demonstrate that methods that neglect the full causal structure may be discarding important dynamical information in a time series.
神经数据分析越来越多地纳入因果信息来研究电路连接性。降维构成了大多数大型多元时间序列分析的基础。在这里，我们提出了一种基于多 taper 的新分解方法，用于随机多元时间序列，该方法作用于所有滞后时间序列的协方差（），这与仅使用零滞后信息来分解时间序列（）的标准方法相反。在模拟和神经成像示例中，我们都证明了忽略完整因果结构的方法可能会丢弃时间序列中的重要动态信息。
