Stochastic processes directed by randomized time (Q5288009)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 243354
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Stochastic processes directed by randomized time |
scientific article; zbMATH DE number 243354 |
Statements
Stochastic processes directed by randomized time (English)
0 references
15 December 1993
0 references
subordination
0 references
threshold models
0 references
total positivity
0 references
independent stationary nonnegative increments
0 references
Hougaard process
0 references
gamma distributions
0 references
0.7570735
0 references
0.73245144
0 references
0.73000324
0 references
0.7219843
0 references
0.72162247
0 references
Let \(\{X(t),\;t\geq 0\}\) and \(\{T(t),\;t\geq 0\}\) be independent stochastic processes, \(\{T(t)\}\) having independent stationary nonnegative increments. The process \(\{X(T(t))\), \(t\geq 0\}\) is said to be subordinate to the parent process \(\{X(t)\}\) and directed by the directing process \(\{T(t)\}\). The paper discusses a number of properties and applications of directed processes. When \(\{X(t)\}\) is stochastically smaller than \(\{Y(t)\}\), i.e. \(E\varphi(X(t_ 1),\dots,X(t_ n))\leq E\varphi(Y(t_ 1),\dots,Y(t_ n))\) for any nondecreasing \(\varphi\) and \(0\leq t_ 1\dots<t_ n\), the same is shown to hold for \(\{X(T(t))\}\) and \(\{Y(T(t))\}\) and a similar result is proved for a stochastically increasing parent process. Let \(\{X(t)\}\) have stationary independent increments. The distribution of \(X(T(t))\) is derived, in particular when \(\{X(t)\}\) is Poisson, the directed process then being compound Poisson. When a device has increasing discrete failure rate for shocks, the same is seen to hold when shocks form a subordinated Poisson process. The subordinated Poisson process also occurs in the limit theory of high-level exceedances in random time in an i.i.d. sequence. Poisson and stable processes directed by a Hougaard process are discussed. The increments of a Hougaard process have a three-parameter distribution that includes the stable, inverse Gaussian and gamma distributions, see \textit{P. Hougaard} [Biometrika 73, 387-396 (1986; Zbl 0603.62015)].
0 references