Asymmetry of cross-correlations between intra-day and overnight volatilities

Abstract

We point out a stunning time asymmetry in the short time cross correlations between intra-day and overnight volatilities (absolute values of log-returns of stock prices). While overnight volatility is significantly (and positively) correlated with the intra-day volatility during the following day (allowing thus non-trivial predictions), it is much less correlated with the intra-day volatility during the preceding day. While the effect is not unexpected in view of previous observations, its robustness and extreme simplicity are remarkable.

Dynamic relationship analysis between NAFTA stock markets using nonlinear, nonparametric, non-stationary methods

“…Stock market prices are not necessarily stationary and normally distributed, so the conventional Pearson correlation is not the most suitable estimator for measuring links between stock market returns as it may lead to a biased estimation [83,84]. We therefore use Spearman rank-order correlation that does not require variables with normal distribution (Gaussian), is based on the ranked values for each variable rather than the raw data and is known to be much more robust than the Pearson correlation, which means monotonic relationships, for instance nonlinear associations [65,82,84]. Spearman's rank correlation coefficient can be estimated as follows [52]:…”

A multi-resolution and multivariate analysis of the dynamic relationships between crude oil and petroleum-product prices

“…The first runs from 16/06/2006 to 31/12/2008 and covers the onset of the global financial crisis; the second runs from 06/01/2008 to 31/05/2009 and its main feature is that it also covers the global financial crisis, with low crude oil prices (this is one of the time intervals where crude oil prices are lowest); the third sub-period runs from 01/01/2009 to 31/12/2014 and covers the period with the lowest degree of volatility; and the last runs from 01/01/2015 and 16/02/2017 and its main characteristic is the presence of tight oil production and low crude oil prices (as in the second sub-period). Since crude oil and products prices are not necessarily stationary and normally distributed (see Table 1 and Section 4.1) [25,64] , the conventional Pearson correlation is not the most suitable estimator for measuring the potential relationship between crude and product prices, as it may lead to a biased estimation [25,65]. We therefore used the Spearman rank-order correlation, which does not demand variables with normal distribution, is based on the ranked values for each variable rather than the raw data and is known to be much more robust (than the Pearson correlation), which means monotonic relationships (for instance, non-linear associations).…”