Stock Returns Versus Trading Volume: Is the Correspondence More General?

Abstract

This paper presents a quantitative analysis of the relationship between the stock market returns and corresponding trading volumes using highfrequency data from the Polish stock market. First, for stocks that were traded for sufficiently long period of time, we study the return and volume distributions and identify their consistency with the power-law functions. We find that, for majority of stocks, the scaling exponents of both distributions are systematically related by about a factor of 2 with the ones for the returns being larger. Second, we study the empirical price impact of trades of a given volume and find that this impact can be well described by a square-root dependence: r(V ) ∼ V 1/2 . We conclude that the properties of data from the Polish market resemble those reported in literature concerning certain mature markets.

Unveiling correlations between financial variables and topological metrics of trading networks: Evidence from a stock and its warrant

“…We also find that the distribution of the stock returns is almost symmetric, while the distribution of the warrant returns is left-skewed. The fact that the kurtosis of the financial variables are greater than 3 indicating that their distributions have fat tails, which are consistent with previous works on stocks and warrants for returns and volatility [19][20][21][22], intertrade [23][24][25], and trading volume [26]. volatility of the warrant is 0.9520 in the last window, which is also truncated in plot (b) for better visibility.…”

“…A trading network with higher centralization has more dominating buyers and such kind of network structures are resulted from one or a few large market orders, which also leads to positive returns [27,28,22]. Hence, the price is more likely to increase consecutively and results in a large positive return.…”

Detrended cross-correlations between returns, volatility, trading activity, and volume traded for the stock market companies

“…As regards its origin, there is no decisive conclusion yet. It might well be a consequence of a relation between volume and trading activity (the number of transactions in a unit time interval), between volume and transaction size, or an effect of the fluctuations in order imbalance [4][5][6][7][8][9][10][11].…”

Interplay between Endogenous and Exogenous Fluctuations in Financial Markets

“…This form is inspired by the empirical analysis [55][56][57][58], where the trading activity is proportional to the square of absolute return (thus α = 2). Note that the given form depends on the long-term component of return in the proposed model, x = 1−n f n f .…”

“…[26] we already adjusted parameters of given model to the empirical data of NYSE and FOREX. Good agreement was achieved for the set of parameters: δ = 1/390 trading day = 3.69 min which is equivalent to 1 NYSE trading minute, ε cf = 1.1 and ε f c = 3, ε cc = 3, H = 1000 adjusts the PSDs of the empirical and model time series, a 0 = 1 and a τ = 0.7 are the empirical parameters defining the sensitivity of market returns and trading activity to the populations of agent states, α = 2 is selected on the basis of empirical analysis [55][56][57][58], and h = 0.3 × 10 −8 s −1 is the main time-scale parameter that adjusts the model to fit the real time-scale. It was shown in [26] that the proposed model of the financial markets with the same set of parameters reproduces PDF and PSD of absolute return and statistics of volatility return intervals for all assets analyzed from the NYSE and FOREX markets with return definition times ∆ ranging from one minute to one month.…”