Multiple breaks detection in financial interval-valued time series

Multiple structural breaks detection for Interval-Valued Time Series (IVTS) is undoubtedly relevant under practical perspectives and challenging the point of view analysis expert systems. In this respect, financial time series usually show high variability outliers; moreover, they often exhibit property being frequency nature; thus, it naturally advisable to consider them as IVTS type a given unit. Despite relevance, scarce effort has been spent by scholars apply methodological advancements in crucial environment series. This paper contributes fill gap. It employs Atheoretical Regression Trees framework – very recent tool that able automatically locale multiple occurring unknown dates stock prices. Such procedure estimate an efficient way considered series; at same time, keeps into account main characteristics intervals describing IVTS. For our purposes, we adopt theoretical proposal reading daily prices whose bounds are defined through closing Empirical experiments on American International Group have experienced past validate model usefulness proposed procedure.