Bridging stylized facts in finance and data non-stationarities

Abstract

Employing a recent technique which allows the representation of nonstationary data by means of a juxtaposition of locally stationary paths of different length, we introduce a comprehensive analysis of the key observables in a financial market: the trading volume and the price fluctuations. From the segmentation procedure we are able to introduce a quantitative description of statistical features of these two quantities, which are often named stylized facts, namely the tails of the distribution of trading volume and price fluctuations and a dynamics compatible with the U-shaped profile of the volume in a trading section and the slow decay of the autocorrelation function. The segmentation of the trading volume series provides evidence of slow evolution of the fluctuating parameters of each patch, pointing to the mixing scenario. Assuming that long-term features are the outcome of a statistical mixture of simple local forms, we test and compare different probability density functions to provide the long-term distribution of the trading volume, concluding that the log-normal gives the best agreement with the empirical distribution. Moreover, the segmentation of the magnitude price fluctuations are quite different from the results for the trading volume, indicating that changes in the statistics of price fluctuations occur at a faster scale than in the case of trading volume. PACS. 05.10.-a Computational methods in statistical physics and nonlinear dynamics -05.45.Tp Time series analysis -89.65.Gh Economics; econophysics, financial markets, business and management

Uncovering the evolution of nonstationary stochastic variables: The example of asset volume-price fluctuations

“…We have shown that these distributions are non-stationary, in the sense that the parameters charactering the distribution are themselves stochastic variables. In order to find the best fit for the volume-price distribution we tested four bi-parametric models commonly used in modelling the price of financial assets [16], namely: the Γ-distribution, inverse Γ-distribution, log-normal distribution and the Weibull distribution.…”

Intraday Seasonalities and Nonstationarity of Trading Volume in Financial Markets: Individual and Cross-Sectional Features

“…On average, we have 126 days per semester. We could have used other segmentation approaches, e.g., the method introduced in [29] that was already applied in the analysis of financial quantities [30]; however, it is likely that its application would give patches of uneven duration and thus the existence of rather small intervals yielding insufficient statistics for them. Alternatively, we can average over companies as well.…”

From human mobility to renewable energies

“…Even when considering only small samples of data sets representing connected agents, analyzing the correlations between them quickly leads to complex problems [15,16]. Further examples range from uncovering the correct statistical description of financial time series [17], as far as they exist in a non-stationary system [18], to devising trading strategies which could mitigate sudden crashes [19], to finding regulatory measures which can stabilize the markets [13].…”