Viruses are small particles of RNA (Ribonucleic Acid) or DNA (DeoxyRibonucleic Acid) wrapped in a protein coat, which can be crystallized into a variety of regular geometric, often polyhedral, shapes. They are much smaller than bacteria, and are capable of passage through filters designed to, arrest these. Among the numerous viruses, bacteriophages (or bacterial viruses), called phages for short, have been the subject of much concentrated study. Over the past twenty years in particular, their structure, parasitic cycle, and most recently part of their genetic mapping, have been elucidated; the isolated DNA strand used by phages as their genetic information carrier, makes them eminently suitable in investigations on the molecular basis of life.

This paper applies the theory of semi-Markov processes to the construction of a stochastic model for interpreting data obtained from clinical trials. The model characterizes the patient as being in one of a finite number of states at any given time with an arbitrary probability distribution to describe the length of stay in a state. Transitions between states are assumed to be chosen according to a stationary finite Markov chain.

Other attempts have been made to develop stochastic models of clinical trials. However, these have all been essentially Markovian with constant transition probabilities which implies that the distribution of time spent during a visit to a state is exponential (or geometric for discrete Markov chains). Markov models need also to assume that the transitions in the state of a patient depend only on absolute time whereas the semi-Markov model assumes that transitions depend on time relative to a patient. Thus the models are applicable to degenerative diseases (cancer, acute leukemia), while Markov models with time dependent transition probabilities are applicable to colds and epidemic diseases. In this paper the Laplace transforms are obtained for (i) probability of being in a state at time t, (ii) probability distribution to reach absorption state and (iii) the probability distribution of the first passage times to go from initial states to transient or absorbing states, transient to transient, and transient to absorbing. The model is applied to a clinical study of acute leukemia in which patients have been treated with methotrexate and 6-mercaptopurine. The agreement between the data and the model is very good.

Considered are two mutually independent recurrent processes each consisting of a time series of unitary stimuli. The durations of the intervals between the stimuli in each series are independent of each other and identically distributed with probability density functions φ (t) and ψ (t). Every stimulus of the ψ (t) process annihilates the next stimulus of the φ (t) process. The probability density function of the intervals of the transformed φ (t) process is derived for the case where either the φ (t) or the ψ (t) process is Poisson.

Several authors have recently discussed the asymptotic properties of stochastic populations which diffuse randomly throughout a given region. Sevast'yanov ([8], [9]) has investigated the extinction probability of a Markovian population in a compact region with an absorbing boundary, his analysis being in terms of “generation times”. Adke and Moyal have considered the spatial dispersion of a population which multiplies according to a simple time-dependent birth-and-death process and undergoes Gaussian diffusion on the real line ([2] and [3]) or on a finite interval with reflecting boundaries [1]. A serious limitation in Adke and Moyal's asymptotic results is that they are conditional upon a finite number of survivors. Moyal [7] has also obtained some basic formulae for a Markovian population diffusing through a general space.

First consider two particles diffusing on the same line, with positions at time t given by y1(t) and y2(t) respectively. We suppose that they cannot pass one another, so that if initially y1(0) < y2(0), then . As long as the two particles are not in contact, we suppose that each moves, independently of the other, according to the Wiener process. We must also prescribe what happens when collisions occur. There seems to be no unique natural way to do this, and we shall adopt a method recommended by its mathematical convenience. However, we shall see that there are two possible motivations for the prescription, one in terms of random walks, the other in terms of a more mechanical model.

Through the use of a zone mapping transformation, geometrical probability is applied to the problem of measuring the area of a planar figure by counting the points covered when the figure is placed at random on a regular point lattice. An expression for the variance σ2 is obtained as a function of radius for the case of a circle. This expression is equivalent to a more general formula for σ2 derived earlier by Kendall and Rankin using analytical methods. Computer calculations of σ are given for two special cases and the results are compared with those predicted by alternative asymptotic formulae also due to Kendall and Rankin. Some examples of the use of zone mapping to treat asymmetric figures, not subject to analytical methods, are described.

In the Ehrenfest model with continuous time one considers two urns and N balls distributed in the urns. The system is said to be in state i if there are i balls in urn I, N − i balls in urn II. Events occur at random times and the time intervals T between successive events are independent random variables all with the same negative exponential distribution

When an event occurs a ball is chosen at random (each of the N balls has probability 1/N to be chosen), removed from its urn, and then placed in urn I with probability p, in urn II with probability q = 1 − p, (0 < p < 1).

In this paper we present a method for obtaining explicit results for some two-sided boundary problems involving sums of independent random variables with alternating distributions. We apply the method to finding the first passage time to either one of two finite barriers, and to some situations arising in queueing and dam theory. The results can be expressed in terms of a finite sum of simple repeated integrals (or sums) of known functions (cf. formulae (3.6)– (3.11)).

A particle that is located at the position (0, 0) in the (x, y) plane at the moment x = 0 is subjected to random impulses at the moments of time x = 1, 2, …, a + b. As the result of each impulse, the particle may be displaced either by a positive or by a negative step with respect to the y-axis; it is given that the particle is restricted by the condition that it has exactly a positive and b negative displacements. Hence in the (x, y) plane the path of the particle at each impulse is depicted by a shift of a unit to the right and by one unit upward or downward; that is the particle's path starts at (0, 0) and terminates at (a + b, a – b) after a + b steps. We assume that all possible different paths are equally probable, and assume also, without loss of generality, that a ≧ b.

A previous paper, hereafter referred to as “I”, reviewed the structure of a class of “regular” diffusion and birth-death processes in one dimension. Such processes were seen to have a basic simplicity manifesting itself in two ways: (1) The ergodic processes have an ergodic distribution that one may write down at once; (2) All such processes have transition functions for which a simple spectral representation is available. The simple spectral structure affords theoretical insight into the structure of related passage time densities, and numerical techniques for their evaluation. In particular, a variational principle is available for estimating the principal eigenvalues and eigenvectors.

We shall consider a many server (multiple channels in parallel) queueing process in which customers arrive at the queue according to a Poisson process. The service times are assumed to be independent and exponentially distributed. As usual, the service times are independent of the arrival process. We assume that no server is idle if there is a customer waiting, but that otherwise the service discipline is arbitrary. If there are n servers and we let Xn(t) denote the number of customers waiting or being served at time t, then it is well known that Xn(t) is a birth and death process with stationary transition probabilities. A very comprehensive analysis of this many server queue from the point of view of birth and death processes has been carried out by Karlin and McGregor [5].

We consider a queueing system in which arrivals occur at times , and after every kth arrival a servicing of k arrivals is begun. We assume that the number of servers is infinite. Initially, at t0= 0, the system is empty and the arrival process {tn} is about to start. The batch service times are independently and identically distributed with distribution function No assumption is made about the process {tn}.

Suppose that we observe an infinite realisation of a point process Π, which is a superposition of a number of mutually independent recurrent point processes Πi, such that

Let us consider a stream of homogeneous random events which we call signals and let us denote by the probability that exactly k signals will appear in the time interval . We assume that the stream is not empty, i.e. the probability P0(a, t) is not identically equal to 1.

In a previous paper [2] the author has studied the single-server queue with non-homogeneous Poisson input and general service time, with particular emphasis on the case when the parameter of the Poisson input is of the form

Transition distribution functions (d.f.) of the stochastic process u + t − X(t), where X(t) has a compound Poisson distribution, are used to derive explicit results for the transition d.f.s of the waiting time processes in the queueing systems M/G/1 and GI/M/1.

Gani (1955) finds the form of the multinomial probabilities which admit a sufficient estimator for the single parameter θ to be where λ θ is a positive function of θ, the Ki, are constants and the αi are nonnegative constants.