Showing papers on "Stochastic simulation published in 1984"
TL;DR: A short algorithm for generating random variates with log-concave density onR and known mode in average number of operations independent off and included in this class are the normal, gamma, Weibull, beta and exponential power, logistic, hyperbolic secant and extreme value distributions.
Abstract: We present a short algorithm for generating random variates with log-concave densityf onR and known mode in average number of operations independent off Included in this class are the normal, gamma, Weibull, beta and exponential power (all with shape parameters at least 1), logistic, hyperbolic secant and extreme value distributions The algorithm merely requires the presence of a uniform [0, 1] random number generator and a subprogram for computingf It can be implemented in about 10 lines of FORTRAN code
131 citations
24 Oct 1984
TL;DR: The authors showed that quasi-random sequences can be used in place of truly random ones for applications such as seeds for pseudo-random number generators, randomizing algorithms, and stochastic simulation experiments.
Abstract: Several applications require truly random bit sequences, whereas physical sources of randomness are at best imperfect. We consider a general model for these slightly-random sources (e,g. zener diodes), and show how to convert their output into 'random looking ' sequences, which we call quasi -random. We show that quasi-random sequences are indistinguishable from truly random ones in a strong sense. This enables us to prove that quasi-random sequences can be used in place of truly random ones for applications such as seeds for pseudo-random number generators, randomizing algorithms, and stochastic simulation experiments.
117 citations
TL;DR: In this article, a diffusion equation including source terms, representing randomly distributed sources and sinks is considered, and the eigenvalue problem is equivalent to that of the quantum mechanical motion of electrons in random fields.
Abstract: A diffusion equation including source terms, representing randomly distributed sources and sinks is considered. For quasilinear growth rates the eigenvalue problem is equivalent to that of the quantum mechanical motion of electrons in random fields. Correspondingly there exist localized and extended density distributions dependent on the statistics of the random field and on the dimension of the space. Besides applications in physics (nonequilibrium processes in pumped disordered solid materials) a new evolution model is discussed which considers evolution as hill climbing in a random landscape.
72 citations
TL;DR: In this paper, an asymptotic distribution theory for the state estimate from a Kalman filter in the absence of the usual Gaussian assumption is presented, where the stability properties of the state transition matrix play a key role in the distribution theory.
Abstract: An asymptotic distribution theory for the state estimate from a Kalman filter in the absence of the usual Gaussian assumption is presented It is found that the stability properties of the state transition matrix playa key role in the distribution theory Specifically, when the state equation is neutrally stable (ie, borderline stable-unstable) the state estimate is asymptotically normal when the random terms in the model have arbitrary distributions This case includes the popular random walk state equation However, when the state equation is either stable or unstable, at least some of the random terms in the model must be normally distributed beyond some finite time before the state estimate is asymptotically normal
57 citations
TL;DR: In this paper, a formalism for sensitivity analysis of stochastic models describing fluctuation phenomena in chemically reacting systems is developed, which is not restricted to chemical kinetics and can be used to analyze any model of a physical system whose state variables obey stochiastic differential equations with white noise.
Abstract: A formalism for sensitivity analysis of stochastic models describing fluctuation phenomena in chemically reacting systems is developed The method is not restricted to chemical kinetics and can be used to analyze any model of a physical system whose state variables obey stochastic differential equations with white noise Expressions for the sensitivity coefficients and densities are obtained These expressions are suitable for direct evaluation by means of a stochastic simulation in a computer The relationship between these quantities and the response functions studied in statistical mechanics is discussed
27 citations
23 citations
01 Aug 1984
TL;DR: The paper illustrates different applications of stochastic simulation techniques to construction activity by describing the computer modelling system that has been developed and the computational methods used to investigate the models.
Abstract: The paper illustrates different applications of stochastic simulation techniques to construction activity. A brief description is provided of discrete-event digital simulation as applied to the construction management process. A historic review of modelling in construction precedes an evaluation of the forms of computer languages and their applicability to construction simulation models. The main requirements for such a language are discussed. The computer modelling system that has been developed is described in the conceptual terms of the symbols that are used to build a schematic representative model in the form of a diagram. Two examples of construction are then described: a simple excavation operation used to validate a model, and a sensitivity analysis of a concrete mixing and distribution system. Schematic diagrams are constructed for both models and example output of some production features for each is presented. Example output from two further models is presented to illustrate further uses of simulation in design and as a basis for contract variations. The computational methods used to investigate the models are described and a general indication of processing time for one of the example models is provided. The paper concludes with an indication of areas of current research. (Author/TRRL)
20 citations
TL;DR: In this article, the authors consider iterating functions that consist of the identity map $[f(x)=x] plus a random function, and investigate the behavior of the entire iterated function.
Abstract: We consider iterating functions that consist of the identity map $[f(x)=x]$ plus a random function. The random function in every iteration is different and has a mean value of zero. We investigate the behavior of the entire iterated function. It is demonstrated that there are three distinct classes of random functions that generate three "phases" of the iterated function. These phases show universal properties independent of the precise form of the added random function. The physical interpretation of the model in terms of aggregation is discussed and an application of the above ideas is made to the problem of particles in a random potential that is varying in space and time.
16 citations
01 Jan 1984
13 citations
TL;DR: In this article, the optimality equation of the best-choice problem with a random number of objects having a known distribution is explicitly solved under conditions on the distribution, which relate to the condition for an OLA policy to be optimal in Markov decision processes.
Abstract: This paper considers the best-choice problem with a random number of objects having a known distribution. The optimality equation of the problem reduces to an integral equation by a scaling limit. The equation is explicitly solved under conditions on the distribution, which relate to the condition for an OLA policy to be optimal in Markov decision processes. This technique is then applied to three different versions of the problem and an exact value for the asymptotic optimal strategy is found.
01 Jan 1984
TL;DR: In this paper, the authors introduce models for which two sets of spacings, constructed from two independent identically distributed (i.i.d.) random variables, can be stochastically ordered.
Abstract: Spacings (that is, the differences between successive order statistics) are useful in various applications in statistics. Many properties of the spacing are known when the spacings are constructed from a collection of independent identically distributed (i.i.d.) random variables. In this paper we study the spacings constructed from not necessarily i.i.d. random variables. We introduce models for which two sets of spacings, constructed from two sets of dependent random variables, can be stochastically ordered. Various examples will be given and applications for goodness-of-fit tests, tests for independence, density estimation and tests for outliers will be discussed.
01 Jan 1984
TL;DR: All phases of input data analysis are covered, including data collection, choosing a modeling distribution, estimating parameters, and goodness-of-fit testing.
Abstract: This paper reviews perspectives and methods for specifying distribution and process forms and parameters from which observations are drawn to drive a stochastic simulation. All phases of input data analysis are covered, including data collection, choosing a modeling distribution, estimating parameters, and goodness-of-fit testing. A discussion is also presented concerning the debate over the desirability of fitting “standard” distributions to data as opposed to using a direct empirical distribution. Available software packages with a simulation orientation are also described.
TL;DR: In this article, the authors deal with charateristics functionals of the solutions of a wide class of stochastic differential equations describing random physical processes, focusing on the cases where an excitation process has the form of a sequence of randomly arriving impulses.
Abstract: This paper deals with charateristics functionals of the solutions of a wide class of stochastic differential equations describing random physical processes. First, a construction of the characteristics functional is shown in a general case of the stochastic evolution equation in Hilbert space. Then making use of the construction presented, the analysis of three random physical processes is performed: the random harmonic oscillator, random wave process and random heat conduction. Special attention in these applications is focused on the cases where an excitation process has the form of a sequence of randomly arriving impulses.
TL;DR: In this article, the transient solution of a class of non-autonomous, stochastic differential equations with given random initial conditions is studied and analytical approximated expressions for the first and second-order moments and for the probability density of the solution process are derived.
TL;DR: In this paper, the impact of random number generators in Monte Carlo time series simulation is discussed, using six different random number generator types and a large number of autoregressive time series of order one are generated.
Abstract: In this paper the impact of random number generators in Monte Carlo time series simulation is discussed. Using six different random number generators a large number of autoregressive time series of order one are generated and some important statistics are compared. From this it seems that unless a very poor generator is used, the choice of generator has little or no impact on the conclusions drawn on the basis of a simulation study.
TL;DR: In this paper, a dynamic model of daily cash and futures prices for cotton was developed using time series analysis, which was included in a recursive Monte Carlo simulation model and validated with a stochastic, dynamic simulation of the estimated model over the observation period 1975-1982.
Abstract: A dynamic model of daily cash and futures prices for cotton was developed using time series analysis. The time series model was included in a recursive Monte Carlo simulation model. Validation of the model was performed with a stochastic, dynamic simulation of the estimated model over the observation period 1975-1982 and with a static, deterministic out-of-sample forecast from December 9, 1981 through March 9, 1982. The model was then used to incorporate futures trading strategies into a policy simulation model.
01 Jan 1984
TL;DR: This paper demonstrates the use of uniformization in continuous random variate generation by representing the continuous random variable of interest as the first passage time of a continuous time stochastic process associated with a Poisson process.
Abstract: In the hybrid simulation/analytic modelling using uniformization, it is required that the basic random variables involved in such modelling are uniformizable. This requires the samples for random variables of interest to be generated by the uniformization procedure. In this paper we demonstrate the use of uniformization in continuous random variate generation. Using uniformization we represent the continuous random variable of interest as the first passage time of a continuous time stochastic process associated with a Poisson process. An approach using this result is proposed to generate samples for continuous random variables. A dynamic uniformization algorithm and other extensions of the basic uniformization algorithm are also considered.
01 Jun 1984
TL;DR: In this article, the Von-Neumann rejection approach via the exponential can be modified to become an efficient algorithm for generating a normal and then present a method for generating normal order statistics.
Abstract: : This paper indicates how exponential random variables can be effeciently used in a variety of simulation problems. One of the problems is the simulation of order statistics from a normal population. The authors discuss the general problem of simulating order statistics and then consider the normal case. They start by showing how the Von-Neumann rejection approach via the exponential can be modified to become an efficient algorithm for generating a normal and then present a method for generating normal order statistics. They show how to use the exponential to efficiently simulate random permutations with weights. They consider the problem of simulating a 2-dimensional Poissin process both for a homogeneous and nonhomogeneous Poisson process. (Author)
TL;DR: In this article, a simple stochastic difference equation is introduced to simulate random sampling drift in population genetics, which is naturally obtained from a random collision model. And then, the model is extended to simulate overdominance.
Abstract: At first we introduce a simple stochastic difference equation, to simulate random sampling drift in population genetics, which is naturally obtained from a random collision model. Next, we introduce a random collision model to simulate overdominance model in population genetics. We assume in a time interval °t, a random collision of four particles, which represents overdominant selection, takes place at a certain probability, where a particle corresponds to a gene. We assume that mutation takes place by some rate and assume that every new mutation is different from extant alleles. We estimate mean heterozygosity by our simulation method and compare it with the result obtained by using a stochastic difference equation for overdominance model.
TL;DR: In this paper, a method of stochastic simulation is applied to the Brusselator chemical reactions and used to model the time course of chemical abundances in this system, and statistical ensembles are used to calculate the nonequilibrium probability density and the expectation values of the reactant concentrations.
Abstract: A method of stochastic simulation is applied to the Brusselator chemical reactions and used to model the time course of chemical abundances in this system. Statistical ensembles are used to calculate the nonequilibrium probability density and the expectation values of the reactant concentrations. The time development of the expectation values is found to depend in a novel manner on the thermodynamic size of the system. The first moments of the ensemble do not remain on or near the limit cycle; in contrast to the solutions of the deterministic reaction‐rate equations they spiral inward toward the center of the limit cycle in a manner determined by the thermodynamic size of the system. Previously predicted, this behavior has not heretofore been observed in an actual system or computational model. The evolution is found to exhibit multiple time scales in which there is a fast initial decay proportional to the overall time required for phase decorrelation. Mechanisms for this scaling are discussed. The genera...
TL;DR: Using Ito's theorem, the authors derives second-order formulae for numerical integration of Langevin equations and shows how constrained systems can be handled and hence how stochastic simulation can be applied to U(1), SU(2) and SU(3) variables.
Abstract: Using Ito's theorem the author derives second-order formulae for the numerical integration of Langevin equations. The author also shows how constrained systems can be handled and hence how stochastic simulation can be applied to U(1), SU(2) and SU(3) variables.
TL;DR: A method, that generates 36-bit machine-independent sets of pseudo-random numbers uniformly distributed in the interval (0.0 to 1.0), is proposed, which gives perfect reproducibility regardless of the operating system and/or kind of computer used.
Abstract: A method, that generates 36-bit machine-independent sets of pseudo-random numbers uniformly distributed in the interval (0.0 to 1.0), is proposed. The method has been tested on several computers, including the PDP 11/10 (16 bit), VAX 11/780 (32 bit), CII 10070 (32 bit), and UNIVAC 1110 (36 bit). The pseudo-random sequences. Normal, Log-normal, Binomial, Poisson, and Chi-squared, are calculated from the Uniform distribution. These pseudo-random numbers give perfect reproducibility regardless of the operating system and/or kind of computer used. This is of the first importance in simulation methods, like the Monte Carlo method. The cycle of the pseudo-random sequence has been tested up to 10**7.
Posted Content•
TL;DR: In this article, the authors describe the application to an operational medium-size econometric model, mini-DMS, of methods associating, to deterministic forecasts, a measure of the uncertainty due to the stochastic nature of behavioural equations.
Abstract: This article describes the application to an operational medium-size econometric model, mini-DMS, of methods associating, to deterministic forecasts, a measure of the uncertainty due to the stochastic nature of behavioural equations. After having described the theoretical and practical foundations of the methods, we shal l analyze sequentially the deterministic bias, the uncertainty (standard error) of forecasts and of policy instruments, trying to look at the information from the point of view of the policy maker.
TL;DR: The problem of automaton transformation of a discrete random process into the discrete random distribution with any arbitrarily given accuracy is investigated and a class of “extremally” unstable random signal sources for which the generated random processes can still be transformed into the distribution by an adder modulo n is introduced.
Abstract: In this paper the problem of automaton transformation of a discrete random process into the discrete random distribution (1/ n , 1/ n ,…, 1/ n ) with any arbitrarily given accuracy is investigated. It is supposed that the properties of these input processes are not completely known, i.e., they are generated by an unstable random signal source. A class of “extremally” unstable random signal sources for which the generated random processes can still be transformed into the distribution (1/ n , 1/ n ,…, 1/ n ) by an adder modulo n is introduced. An estimate of the distance between the obtained and the discrete uniform random distributions is also found, dependent on the number of operating steps of the adder, in the case when the input process is described by a finite Markov chain.
TL;DR: In this paper, a random coefficient model in which means of random coefficients are subject to a set of linear stochastic constraints is considered and estimators for the means of coefficients are proposed.
Abstract: A random coefficient model in which means of random coefficients are subject to a set of linear stochastic constraints is considered and estimators for the means of coefficients are proposed. Their asymptotic properties are presented and some remarks on efficiency are placed.
Journal Article•