Automated AI - Equity Market Lead–Lag Prediction Based on Multivariate Time Series

The lead-lag structure among time-series factors is the recognition that in multivariate time-series frameworks, some factor clusters significantly drive improvements in the framework, while various factors follow this development with time lags arises from. In this article, we propose a multivariate framework for the identification of lead-lag bundles in time series. We demonstrate that pairwise lead-lag interactions between time series can be viewed as a directed organization. There is a practical calculation for locating the lead-lag bundle set with significant pairwise imbalance in this instance. We investigate various options for the system's directed network clustering models and pairwise lead lag metric. Daily cost data on actual US values and the constructed generative model of the multivariate lead-lag time series framework are used to test the system. Using the pairwise lead-lag metric and directed tissue clustering computations, we offer a method for identifying lead-lag clusters in multivariate time series without a doubt. We demonstrate that the stationary tissue of pairwise lead-lag interactions between time series can be conceptualized as a directed tissue and that a suitable computation exists for locating the lead-lag groups in this type of tissue that have significant pairwise imbalances. increase.