How does trading volume affect financial return distributions?

Abstract

We assess investors' reaction to new information arrivals in financial markets by examining the relationships between trading volume and the higher moments of returns in 18 international equity and currency markets. Our volume-volatility results support extant information theories and further contribute new evidence of cross market relations between volume and volatility. We also find that the direct impact of volume on the level of negative skewness is less significant for more diversified regional portfolios. Furthermore, the negative interaction between volume and kurtosis can be explained by the differences of opinion in financial markets. We observe stronger interdependence among higher moments in reaction to significant events, but the strength is dampened by trading volume. This result is consistent with trading volume being a source of heteroskedasticity in asset returns.

Investor Attention and Stock Market Activities: New Evidence from Panel Data

“…Thus, searching behavior is time-varying. In addition, we cannot neglect the important role of trading volume on the higher moments of the stock return distribution as suggested by Do et al (2014). In the study the attention-abnormal trading volume relationship has the strongest results with relatively higher numbers of statistically significant coefficients for all samples in our analysis.…”

Learning from SARS: Return and volatility connectedness in COVID-19

“…As shown in Do et al (2013, 2014 ), to incorporate the Diebold and Yilmaz approach in a FIVAR model, the moving coefficient matrix Λ h need to be adjusted with the long memory degree ( d ) as, , where } is a (2 × 2) diagonal matrix, and Φ h is calculated recursively as, . We note that, Φ 0 = Λ 0 = I 2 , and e r is the identity vector with unity as its r th element.…”

Interconnectedness in the Australian National Electricity Market: A Higher‐Moment Analysis*

“…7 FIVAR models can also be estimated using a two-step estimation method, which has commonly been employed in previous studies (e.g., Do et al, 2014;Yip et al, 2017). With the two-step method, the first stage estimates the vector of memory degrees (d) in a multivariate framework such as Shimotsu (2007).…”