• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.825-834
• /
• 2002

Poisson processes are widely used in reliability and survival analysis. In particular, multiple event time data in survival analysis are routinely analyzed by use of Poisson processes. In this paper, we consider large sample properties of nonparametric Bayesian models for Poisson processes. We prove that the posterior distribution of the cumulative intensity function of Poisson processes is consistent under regularity conditions on priors which are Levy processes.

• Journal of the Korean Mathematical Society
• /
• v.35 no.1
• /
• pp.207-224
• /
• 1998

After the review of known results on the connections between selfsimilar processes with independent increments (processes of class L) and selfdecomposable distributions and between semi-selfsimilar processes with independent increments and semi-selfdecomposable distributions, dichotomy of those processes into transient and recurrent is discussed. Due to the lack of stationarity of the increments, transience and recurrence are not expressed by finiteness and infiniteness of mean sojourn times on bound sets. Comparison in transience-recurrence of the Levy process and the process of class L associated with a common distribution of class L is made.

• Journal of the Korean Statistical Society
• /
• v.24 no.1
• /
• pp.65-76
• /
• 1995

In this paper, we show that the result of random upper functions for Levy processes obtained by Joo(1993) can be sharpened under some additional assumption. This is the continuous analogue of result obtained by Griffin and Kuelbs (1989) for sums of i.i.d. random varialbles.

• Journal of the Korean Mathematical Society
• /
• v.35 no.3
• /
• pp.583-611
• /
• 1998

We consider a general class of real-valued Levy processes {X(t), $t\geq0$}, and obtain suitable large deviation results for the empiricals L(t, A) defined by $t^{-1}{\int^t}_01_A$(X(s)ds for t > 0 and a Borel subset A of R. These results are used to obtain the asymptotic behavior of P{Z(t) < a}, where Z(t) = $sup_{u\leqt}\midx(u)\mid$ as $t\longrightarrow\infty$, in terms of the rate function in the large deviation principle. A subclass of these processes is the Feller class: there exist nonrandom functions b(t) and a(t) > 0 such that {(X(t) - b(t))/a(t) : t > 0} is stochastically compact, i.e., each sequence has a weakly convergent subsequence with a nondegenerate limit. The stable processes are in this class, but it is much larger. We consider processes in this class for which b(t) may be taken to be zero. For any t > 0, we consider the renormalized process ${X(u\psi(t))/a(\psi(t)),u\geq0}$, where $\psi$(t) = $t(log log t)^{-1}$, and obtain large deviation probability estimates for $L_{t}(A)$ := $(log log t)^{-1}$${\int_{0}}^{loglogt}1_A$$(X(u\psi(t))/a(\psi(t)))dv$. It turns out that the upper and lower bounds are sharp and depend on the entire compact set of limit laws of {X(t)/a(t)}. The results extend to random walks in the Feller class as well. Earlier results of this nature were obtained by Donsker and Varadhan for symmetric stable processes and by Jain for random walks in the domain of attraction of a stable law.

• Journal of the Korean Mathematical Society
• /
• v.22 no.2
• /
• pp.199-210
• /
• 1985

• Honam Mathematical Journal
• /
• v.12 no.1
• /
• pp.113-123
• /
• 1990

• Journal of the Korean Statistical Society
• /
• v.22 no.1
• /
• pp.93-111
• /
• 1993

Let ${X(t) : t \geq 0}$ be a real-valued stochastics process with stationary independent increments. In this paper, under the condition of stochastic compactness, we obtain appropriate function $\alpha(t)$ and random function $\beta(t)$ such that for some positive finite constant C, lim sup${X(t) - \alpha(t)}/\beta(t) = C$ a.s. both as t tends to zero and infinity.

• Management Science and Financial Engineering
• /
• v.19 no.2
• /
• pp.37-42
• /
• 2013

This paper suggests a numerical method for valuation of European and American options under the two L$\acute{e}$vy Processes, Normal Inverse Gaussian Model and the Variance Gamma model. The method is based on approximation of underlying asset price using a finite-state, time-homogeneous Markov chain. We examine the effectiveness of the proposed method with simulation results, which are compared with those from the existing numerical method, the lattice-based method.

• Journal of the Korean Mathematical Society
• /
• v.36 no.1
• /
• pp.139-157
• /
• 1999

Consider a supercritical Bellman-Harris process evolving from one particle. We superimpose on this process the additional structure of movement. A particle whose parent was at x at its time of birth moves until it dies according to a given Markov process X starting at x. The motions of different particles are assumed independent. In this paper we show that when the movement process X is standard Brownian the proportion of particles with position $\leq${{{{ SQRT { t} }}}} b and age$\leq$a tends with probability 1 to A(a)$\Phi$(b) where A(.) and $\Phi$(.) are the stable age distribution and standard normal distribution, respectively. We also extend this result to the case when the movement process is a Levy process.

• Proceedings of the Korea Air Pollution Research Association Conference
• /
• 2003.05b
• /
• pp.121-122
• /
• 2003

Deposition to the sea surface is one of ozone's principal loss mechanisms (Galbally and Roy, 1980; Levy et al., 1985; Kramm, 1995). However, since complicated physical and chemical processes are involved, large uncertainties remain in evaluating this loss mechanism that need to be better characterized. In this study we attempted to explore possible causes that give rise to large variability of ozone deposition velocity in terms of wind speed and chemical reactivity in the aqueous-phase film. (omitted)