H. E. Roman

Characterization of the streamer regime in dielectric barrier discharges

The streamer regime of a dielectric barrier discharge device is studied by performing a detailed statistical analysis of current-voltage measurements in air. A wide bandwidth Rogowski coil, designed to work down to the nanoseconds time scale, is used to record the discharge current. The temporal structure of the latter is identified and characterized by its probability density distribution as a function of the applied voltage. The results suggest the existence of two discharge regimes, separated by a well defined voltage threshold, reflecting the different behaviors of the microdischarges. The autocorrelations of the discharge signal are evaluated as a function of the applied voltage, indicating the presence of strong correlations at short-time scales (up to the order of 102 ns) and residual correlations at longer times. The latter are shown to be due to the nonstationarity of the discharge process.

Structurally constrained protein evolution: results from a lattice simulation.

Abstract:We simulate the evolution of a protein-like sequence subject to point mutations, imposing conservation of the ground state, thermodynamic stability and fast folding. Our model is aimed at describing neutral evolution of natural proteins. We use a cubic lattice model of the protein structure and test the neutrality conditions by extensive Monte Carlo simulations. We observe that sequence space is traversed by neutral networks, i.e. sets of sequences with the same fold connected by point mutations. Typical pairs of sequences on a neutral network are nearly as different as randomly chosen sequences. The fraction of neutral neighbors has strong sequence to sequence variations, which influence the rate of neutral evolution. In this paper we study the thermodynamic stability of different protein sequences. We relate the high variability of the fraction of neutral mutations to the complex energy landscape within a neutral network, arguing that valleys in this landscape are associated to high values of the neutral mutation rate. We find that when a point mutation produces a sequence with a new ground state, this is likely to have a low stability. Thus we tentatively conjecture that neutral networks of different structures are typically well separated in sequence space. This result indicates that changing significantly a protein structure through a biologically acceptable chain of point mutations is a rare, although possible, event.

Skewness, long-time memory, and non-stationarity: Application to leverage effect in financial time series

We analyze daily log-returns data for a set of 1200 stocks, taken from US stock markets, over a period of 2481 trading days (January 1996–November 2005). We estimate the degree of non-stationarity in daily market volatility employing a polynomial fit, used as a detrending function. We find that the autocorrelation function of absolute detrended log-returns departs strongly from the corresponding original data autocorrelation function, while the observed leverage effect depends only weakly on trends. Such effect is shown to occur when both skewness and long-time memory are simultaneously present. A fractional derivative random walk model is discussed yielding a quantitative agreement with the empirical results.

Fluctuation analysis of proficient and dysgraphic handwriting in children

We analyze handwriting records from several school children with the aim of characterizing the fluctuating behavior of the writing speed. It will be concluded that remarkable differences exist between proficient and dysgraphic handwritings which were unknown so far. It is shown that in the case of proficient handwriting, the variations in handwriting speed are strongly autocorrelated within times corresponding to the completion of a single character or letter, while become uncorrelated at longer times. In the case of dysgraphia, such correlations persist on longer time scales and the autocorrelation function seems to display algebraic time decay, indicating the presence of strong anomalies in the handwriting process. Applications of the results in educational/clinical programs are envisaged.

Wavelet analysis of two-dimensional turbulence in a pure electron plasma

Highly magnetized pure electron plasmas confined in Penning-Malmberg traps allow to perform accurate experiments on effective two-dimensional (2D) fluid dynamics under well-controlled experimental conditions. Detailed experimental investigations of free decaying turbulence in a pure electron plasma using multi-resolution analysis (MRA) are presented. Coherent structures are extracted from the flow using a wavelet-based recursive denoising algorithm with an adaptative self-consistent threshold. The scaling behaviors of measured enstrophy and energy distributions are obtained and compared with recent experiments and theoretical predictions for 2D turbulence.

Fluctuations of company yearly profits vs. scaled revenue: Fat-tail distribution of Lévy type

We analyze annual revenues and earnings data for the 500 largest-revenue U.S. companies during the period 1954-2007. We find that mean year profits are proportional to mean year revenues, exception made for few anomalous years, from which we postulate a linear relation between company expected mean profit and revenue. Mean annual revenues are used to scale both company profits and revenues. Annual profit fluctuations are obtained as difference between actual annual profit and its expected mean value, scaled by a power of the revenue to get a stationary behavior as a function of revenue. We find that profit fluctuations are broadly distributed having approximate power-law tails with a Levy-type exponent

Anomalous scaling of stock price dynamics within ARCH-models

\alpha \simeq 1.7

Enhanced light emission in Si-nanoclusters arrays

, from which we derive the associated break-even probability distribution. The predictions are compared with empirical data.