𝔖 Bobbio Scriptorium
✦   LIBER   ✦

The Asymptotic Loss of Information for Grouped Data

✍ Scribed by Klaus Felsenstein; Klaus Pötzelberger

Publisher
Elsevier Science
Year
1998
Tongue
English
Weight
390 KB
Volume
67
Category
Article
ISSN
0047-259X

No coin nor oath required. For personal study only.

✦ Synopsis

We study the loss of information (measured in terms of the Kullback Leibler distance) caused by observing ``grouped'' data (observing only a discretized version of a continuous random variable). We analyze the asymptotical behaviour of the loss of information as the partition becomes finer. In the case of a univariate observation, we compute the optimal rate of convergence and characterize asymptotically optimal partitions (into intervals). In the multivariate case we derive the asymptotically optimal regular sequences of partitions. Furthermore, we compute the asymptotically optimal transformation of the data, when a sequence of partitions is given. Examples demonstrate the efficiency of the suggested discretizing strategy even for few intervals.

📜 SIMILAR VOLUMES

The Comparison of Two Sets of Grouped Cy
The Comparison of Two Sets of Grouped Cyclic (Circular) Data
✍ Dr. Graham Upton 📂 Article 📅 2007 🏛 John Wiley and Sons 🌐 English ⚖ 245 KB 👁 1 views

## Summry A Monte Carlo procedure is suggested for the problem of judging whether two sets of grouped circular data, possibly differently oriented, have arisen from otherwise equivalent dietributions. The procedure is applied to data relating to t h e months of onset of two illnesses in a group of

A Test of Equality of Survival Distribut
A Test of Equality of Survival Distributions for Two Sample Censored Grouped Survival Data
✍ Dr. Terence J. O'Neill 📂 Article 📅 1985 🏛 John Wiley and Sons 🌐 English ⚖ 454 KB 👁 3 views

A coninion testing problem for a life table or survival date is to test the equality of two survival distributions when the data is both grouped end censored. Several tests have been proposed in the literature which require various assumptions about the censoring distributions. It is shown that if t