Article ID: | iaor1999333 |
Country: | United Kingdom |
Volume: | 29 |
Issue: | 4 |
Start Page Number: | 947 |
End Page Number: | 964 |
Publication Date: | Dec 1997 |
Journal: | Advances in Applied Probability |
Authors: | Yeo Geoffrey F., Stefanov Valeri T. |
The dynamical aspects of single channel gating can be modelled by a Markov renewal process, with states aggregated into two classes corresponding to the receptor channel being open or closed, and with brief sojourns in either class not detected. This paper is concerned with the relation between the amount of time, for a given record, in which the channel appears to be open compared to the amount in which it is actually open and the difference in their proportions; this may be used to obtain information on the unobserved actual process from the observed one. Results, with extensions, on exponential families have been applied to obtain relevant generating functions and asymptotic normal distributions, including explicit forms for the parameters. Numerical results are given as illustration in special cases.