LADDER REGENERATIVE EVENTS WITH APPLICATIONS TO DAM MODELS,
Abstract
For a stochastic process (X(t); t greater than or equal to 0) with stationary independent increments, we define an event called a ladder event. The event is a regenerative event in the sense of Kingman. Conditions for the event to be standard or stable are given and its local time properties, limiting properties, and sample function behavior are investigated. It is found that there are two classes of processes with stationary independent increments, for one class the event is standard while for the other it is degenerative. Ladder epochs for X(t) are introduced as first passage times. Their behavior is characterized using the ladder events. The connection between ladder epochs and the supremum functional for X(t) is used to obtain limit theorems for the latter for a large class of processes with stationary independent increments. Ladder events, ladder epochs and the supremum functional are further studies for the special process which is representable as the sum of two independent processes, one, a process with stationary independent increments, and the other, a compound Poisson process. The approach here uses a continuous time extension of Feller's combinatorial lemma. Finally, these results are applied to investigate a general dam model. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1968
- Accession Number
- AD0684859
Entities
People
- Michael Rubinovitch
Organizations
- Cornell University