Abstract : Suppose that two (possibly dependent) point processes are observed simultaneously over a period of time, yielding observations at times A1 < A2 ... < AN(A) for the first process, and at times B1 < B2 < ... BN(B) for the second. Such data arises in many contexts, and it is often of interest to discover and quantify the association between the two processes. Two fields in which this situation occurs are neurophysiology and reliability theory. In this article, we describe and discuss certain graphs, plots, as well as more formal methods that can assess the dependence between point processes. Specifically, these methods indicate whether or not the likelihood of an A-point is increased (decreased) after the occurrence of a B-point. The techniques are illustrated on simulated data. Although bivariate point processes arise in many fields, we emphasize the applications in neurophysiology. Keywords: Reliability theory.
[1]
Hani Doss,et al.
On estimating the dependence between two point processes
,
1989
.
[2]
J. Griffith,et al.
Functional Coupling between Cells in the Visual Cortex of the Unrestrained Cat
,
1963,
Nature.
[3]
D. Brillinger.
Estimation of the Second-Order Intensities of a Bivariate Stationary Point Process
,
1976
.
[4]
P. A. W. Lewis,et al.
Multivariate point processes
,
2018,
Point Processes.
[5]
D. R. Cox,et al.
On the Estimation of the Intensity Function of a Stationary Point Process
,
1965
.
[6]
M. R. Leadbetter.
On three basic results in the theory of stationary point processes
,
1968
.