Fixed link
0
vote

Persistent Intraday Correlations Create Skews in Daily-Scale Distributions Revisited: Behaviors of High-Magnitude Fluctuations

T. J. Mazurek

posted on 04 May 2018

download (177 views, 158 download, 0 comments)

This work began with an earlier study that showed the extreme persistence of intraday correlations in order-flow and price-mobility creates positively skewed daily-scale distributions of bull-bear sentiment and price-change, respectively.  The earlier work also dealt with the all-magnitude skew excesses as one of each distribution’s summary parameters and compared scaling behaviors of all corresponding parameters of price-change and sentiment.  The present study extends this prior work by determining the price-mobility and order-flow persistent correlations for high magnitude price-changes.  This study shows that the bias in daily occurrence probabilities at high-magnitudes determined from price-mobility correlations favors negative price-changes, reversing the positive bias of the all-magnitude probabilities. This produces a negative daily price-mobility skew that creates an equal negative daily price-change distribution’s skew excess at high magnitudes.  The behavior of high-magnitude order-flow correlations parallels the behavior of price-mobility described above to create a similar negative sentiment distribution skew at high-magnitudes.  The present study then compares behaviors with increasing fluctuation magnitudes of price-change and sentiment skews on the daily and seven-day scale.  These scaling comparisons show that the daily negative skew strength at high-magnitudes is amplified at successively higher scales as was the daily positive all-magnitude skew strength in the earlier work.  Such amplifications sharply contrast with behaviors of skew strengths for independent identical distributions which must decay with increases in scale.