arXiv subjects

Erik Aurell

Publications and source records attributed to Erik Aurell.

At least 109 records · Page 6Linked to original sources

arXiv2005

Optimal hedging of Derivatives with transaction costs

We investigate the optimal strategy over a finite time horizon for a portfolio of stock and bond and a derivative in an multiplicative Markovian market model with transaction costs (friction). The optimization problem is solved by a Hamilton-Bellman-Jacobi equation, which by the verification theorem has well-behaved solutions if certain conditions on a potential are satisfied. In the case at hand, these conditions simply imply arbitrage-free ("Black-Scholes") pricing of the derivative. While pricing is hence not changed by friction allow a portfolio to fluctuate around a delta hedge. In the limit of weak friction, we determine the optimal control to essentially be of two parts: a strong control, which tries to bring the stock-and-derivative portfolio towards a Black-Scholes delta hedge; and a weak control, which moves the portfolio by adding or subtracting a Black-Scholes hedge. For simplicity we assume growth-optimal investment criteria and quadratic friction.

physics.soc-ph↗

arXiv2005

"Locally homogeneous turbulence" Is it an inconsistent framework?

In his first 1941 paper Kolmogorov assumed that the velocity has increments which are homogeneous and independent of the velocity at a suitable reference point. This assumption of local homogeneity is consistent with the nonlinear dynamics only in an asymptotic sense when the reference point is far away. The inconsistency is illustrated numerically using the Burgers equation. Kolmogorov's derivation of the four-fifths law for the third-order structure function and its anisotropic generalization are actually valid only for homogeneous turbulence, but a local version due to Duchon and Robert still holds. A Kolomogorov--Landau approach is proposed to handle the effect of fluctuations in the large-scale velocity on small-scale statistical properties; it is is only a mild extension of the 1941 theory and does not incorporate intermittency effects.

nlin.CD↗

arXiv2005

A Statistical Theory of Chord under Churn

Most earlier studies of Distributed Hash Tables (DHTs) under churn have either depended on simulations as the primary investigation tool, or on establishing bounds for DHTs to function. In this paper, we present a complete analytical study of churn using a master-equation-based approach, used traditionally in non-equilibrium statistical mechanics to describe steady-state or transient phenomena. Simulations are used to verify all theoretical predictions. We demonstrate the application of our methodology to the Chord system. For any rate of churn and stabilization rates, and any system size, we accurately predict the fraction of failed or incorrect successor and finger pointers and show how we can use these quantities to predict the performance and consistency of lookups under churn. We also discuss briefly how churn may actually be of different 'types' and the implications this will have for the functioning of DHTs in general.

cs.NI↗

arXiv2005

Comparing Beliefs, Surveys and Random Walks

Survey propagation is a powerful technique from statistical physics that has been applied to solve the 3-SAT problem both in principle and in practice. We give, using only probability arguments, a common derivation of survey propagation, belief propagation and several interesting hybrid methods. We then present numerical experiments which use WSAT (a widely used random-walk based SAT solver) to quantify the complexity of the 3-SAT formulae as a function of their parameters, both as randomly generated and after simplification, guided by survey propagation. Some properties of WSAT which have not previously been reported make it an ideal tool for this purpose -- its mean cost is proportional to the number of variables in the formula (at a fixed ratio of clauses to variables) in the easy-SAT regime and slightly beyond, and its behavior in the hard-SAT regime appears to reflect the underlying structure of the solution space that has been predicted by replica symmetry-breaking arguments. An analysis of the tradeoffs between the various methods of search for satisfying assignments shows WSAT to be far more powerful that has been appreciated, and suggests some interesting new directions for practical algorithm development.

cond-mat.stat-mech↗

arXiv2004

The global picture of self-similar and not self-similar decay in Burgers Turbulence

This paper continue earlier investigations on the decay of Burgers turbulence in one dimension from Gaussian random initial conditions of the power-law spectral type $E_0(k)\sim|k|^n$. Depending on the power $n$, different characteristic regions are distinguished. The main focus of this paper is to delineate the regions in wave-number $k$ and time $t$ in which self-similarity can (and cannot) be observed, taking into account small-$k$ and large-$k$ cutoffs. The evolution of the spectrum can be inferred using physical arguments describing the competition between the initial spectrum and the new frequencies generated by the dynamics. For large wavenumbers, we always have $k^{-2}$ region, associated to the shocks. When $n$ is less than one, the large-scale part of the spectrum is preserved in time and the global evolution is self-similar, so that scaling arguments perfectly predict the behavior in time of the energy and of the integral scale. If $n$ is larger than two, the spectrum tends for long times to a universal scaling form independent of the initial conditions, with universal behavior $k^2$ at small wavenumbers. In the interval $2<n$ the leading behaviour is self-similar, independent of $n$ and with universal behavior $k^2$ at small wavenumber. When $1<n<2$, the spectrum has three scaling regions : first, a $|k|^n$ region at very small $k$\ms1 with a time-independent constant, second, a $k^2$ region at intermediate wavenumbers, finally, the usual $k^{-2}$ region. In the remaining interval, $n<-3$ the small-$k$ cutoff dominates, and $n$ also plays no role. We find also (numerically) the subleading term $\sim k^2$ in the evolution of the spectrum in the interval $-3<n<1$. High-resolution numerical simulations have been performed confirming both scaling predictions and analytical asymptotic theory.

nlin.CD↗

arXiv2003

Evidence for a k^{-5/3} spectrum from the EOLE Lagrangian balloons in the low stratosphere

The EOLE Experiment is revisited to study turbulent processes in the lower stratosphere circulation from a Lagrangian viewpoint and resolve a discrepancy on the slope of the atmospheric energy spectrum between the work of Morel and Larcheveque (1974) and recent studies using aircraft data. Relative dispersion of balloon pairs is studied by calculating the Finite Scale Lyapunov Exponent, an exit time-based technique which is particularly efficient in cases where processes with different spatial scales are interfering. Our main result is to reconciliate the EOLE dataset with recent studies supporting a k^{-5/3} energy spectrum in the range 100-1000 km. Our results also show exponential separation at smaller scale, with characteristic time of order 1 day, and agree with the standard diffusion of about 10^7 m^2/s at large scales. A still open question is the origin of a k^{-5/3} spectrum in the mesoscale range, between 100 and 1000 km.

physics.ao-ph↗

arXiv2002

Self-gravitating systems in a three-dimensional expanding Universe

The non-linear evolution of one-dimensional perturbations in a three-dimensional expanding Universe is considered. A general Lagrangian scheme is derived, and compared to two previously introduced approximate models. These models are simulated with heap-based event-driven numerical procedure, that allows for the study of large systems, averaged over many realizations of random initial conditions. One of the models is shown to be qualitatively, and, in some respects, concerning mass aggregation, quantitatively similar to the adhesion model.

cond-mat.dis-nn↗

arXiv2001

An inventory of Lattice Boltzmann models of multiphase flows

This document reports investigations of models of multiphase flows using Lattice Boltzmann methods. The emphasis is on deriving by Chapman-Enskog techniques the corresponding macroscopic equations. The singular interface (Young-Laplace-Gauss) model is described briefly, with a discussion of its limitations. The diffuse interface theory is discussed in more detail, and shown to lead to the singular interface model in the proper asymptotic limit. The Lattice Boltzmann method is presented in its simplest form appropriate for an ideal gas. Four different Lattice Boltzmann models for non-ideal (multi-phase) isothermal flows are then presented in detail, and the resulting macroscopic equations derived. Partly in contradiction with the published literature, it is found that only one of the models gives physically fully acceptable equations. The form of the equation of state for a multiphase system in the density interval above the coexistance line determines surface tension and interface thickness in the diffuse interface theory. The use of this relation for optimizing a numerical model is discussed. The extension of Lattice Boltzmann methods to the non-isothermal situation is discussed summarily.

cond-mat.soft↗

arXiv2001

A Price Dynamics in Bandwidth Markets for Point-to-point Connections

We simulate a network of N routers and M network users making concurrent point-to-point connections by buying and selling router capacity from each other. The resources need to be acquired in complete sets, but there is only one spot market for each router. In order to describe the internal dynamics of the market, we model the observed prices by N-dimensional Ito-processes. Modeling using stochastic processes is novel in this context of describing interactions between end-users in a system with shared resources, and allows a standard set of mathematical tools to be applied. The derived models can also be used to price contingent claims on network capacity and thus to price complex network services such as quality of service levels, multicast, etc.

cs.NI↗

arXiv2001

Heap-based algorithm for one-dimensional particle systems

A fast algorithm to study one-dimensional self-gravitating systems, and, more generally, systems that are Lagrangian integrable between collisions, is presented. The algorithm is event-driven, and uses a heap-ordered set of predicted future events. In the limit of large number of particles $N$, the operation count is dominated by the cost of reordering the heap after each event, which goes asymptotically as $\log N$. Some applications are discussed in detail.

cond-mat.dis-nn↗

arXiv2000

Stability Puzzles in Phage Lambda

The lysogeny maintenance switch in phage lambda is one of the simplest examples on the molecular level of computation, command and control in a living system. If, following infection of the bacterium E. coli, the virus enters the lysogenic pathway, it represses its developmental functions, and integrates its DNA into the host chromosome. In this state the prophage may be passively replicated for many generations of E. coli. In fact, this repressed state is intrinsically more stable than the gene encoding the repressor. We develop a mathematical formalism to predict the stability of such epigenetic states from affinities of the molecular components. We apply the model to the behavior of recently published mutants at the right operator complex of lambda, and find that the reported stability indicates that the current view of the switch is incomplete. The approach described here should be generally applicable to the stability of expressed states

cond-mat.soft↗

arXiv2000

Drifters dispersion in the Adriatic Sea: Lagrangian data and chaotic model

We analyze characteristics of drifter trajectories from the Adriatic Sea with recently introduced nonlinear dynamics techniques. We discuss how in quasi-enclosed basins, relative dispersion as function of time, a standard analysis tool in this context, may give a distorted picture of the dynamics. We further show that useful information may be obtained by using two related non-asymptotic indicators, the Finite-Scale Lyapunov Exponent (FSLE) and the Lagrangian Structure Function (LSF), which both describe intrinsic physical properties at a given scale. We introduce a simple chaotic model for drifter motion in this system, and show by comparison with the model that Lagrangian dispersion is mainly driven by advection at sub-basin scales until saturation sets in.

chao-dyn↗

arXiv2000

On the dynamics of a self-gravitating medium with random and non-random initial conditions

The dynamics of a one-dimensional self-gravitating medium, with initial density almost uniform is studied. Numerical experiments are performed with ordered and with Gaussian random initial conditions. The phase space portraits are shown to be qualitatively similar to shock waves, in particular with initial conditions of Brownian type. The PDF of the mass distribution is investigated.

chao-dyn↗

arXiv1999

Growth Optimal Investment and Pricing of Derivatives

We introduce a criterion how to price derivatives in incomplete markets, based on the theory of growth optimal strategy in repeated multiplicative games. We present reasons why these growth-optimal strategies should be particularly relevant to the problem of pricing derivatives. We compare our result with other alternative pricing procedures in the literature, and discuss the limits of validity of the lognormal approximation. We also generalize the pricing method to a market with correlated stocks. The expected estimation error of the optimal investment fraction is derived in a closed form, and its validity is checked with a small-scale empirical test.

cond-mat.stat-mech↗

arXiv1999

Financial Friction and Multiplicative Markov Market Game

We study long-term growth-optimal strategies on a simple market with linear proportional transaction costs. We show that several problems of this sort can be solved in closed form, and explicit the non-analytic dependance of optimal strategies and expected frictional losses of the friction parameter. We present one derivation in terms of invariant measures of drift-diffusion processes (Fokker- Planck approach), and one derivation using the Hamilton-Jacobi-Bellman equation of optimal control theory. We also show that a significant part of the results can be derived without computation by a kind of dimensional analysis. We comment on the extension of the method to other sources of uncertainty, and discuss what conclusions can be drawn about the growth-optimal criterion as such.

cond-mat.stat-mech↗

arXiv1998

Risk-return arguments applied to options with trading costs

We study the problem of option pricing and hedging strategies within the frame-work of risk-return arguments. An economic agent is described by a utility function that depends on profit (an expected value) and risk (a variance). In the ideal case without transaction costs the optimal strategy for any given agent is found as the explicit solution of a constrained optimization problem. Transaction costs are taken into account on a perturbative way. A rational option price, in a world with only these agents, is then determined by considering the points of view of the buyer and the writer of the option. Price and strategy are determined to first order in the transaction costs.

cond-mat.stat-mech↗

arXiv1997

Torsion and electron motion in Quantum Dots with crystal lattice dislocations

The motion of a conducting electron in a quantum dot with one or several dislocations in the underlying crystal lattice is considered in the continuum picture, where dislocations are represented by torsion of space. The possible effects of torsion are investigated on the levels of classical motion, non-relativistic quantum motion, and spin-torsion coupling terms derivable in the non-relativistic limit of generalizations of the Dirac equation in a space with torsion. Finally, phenomenological spin-torsion couplings analogous to Pauli terms are considered in the non-relativistic equations. Different prescriptions of classical and non-relativistic quantum motion in a space with torsion are shown to give effects that should in principle be observable. Semi-classical arguments are presented to show that torsion is not relevant for the classical motion of the center of a wave packet. The correct semi-classical limit can instead be described as classical trajectories in a Hamiltonian given by the band energy. In the special case of a spherically symmetric band this motion reduces to straight lines, independently of local crystal orientations. By dimensional analysis the coupling constants of the possible spin-torsion interactions are postulated to be proportional to a combination of the effective mass of the electron, the lattice constant and $\hbar$. The level-splitting is then very small with transition frequencies on the order of 1 kHz or smaller.

cond-mat.mes-hall↗

arXiv1996

Numerical Proof of Self-Similarity in Burgers' Turbulence

We study the statistical properties of solutions to Burgers' equation, $v_t + vv_x = νv_{xx}$, for large times, when the initial velocity and its potential are stationary Gaussian processes. The initial power spectral density at small wave numbers follows a steep power-law $E_0(k) \sim |k|^n$ where the exponent $n$ is greater than two. We compare results of numerical simulations with dimensional predictions, and with asymptotic analytical theory. The theory predicts self-similarity of statistical characteristics of the turbulence, and also leads to a logarithmic correction to the law of energy decay in comparison with dimensional analysis. We confirm numerically the existence of self-similarity for the power spectral density, and the existence of a logarithmic correction to the dimensional predictions.

patt-sol↗