The literature oers many formulas for estimating the mean and standard deviation of a subjective probability distribution (a well-known example is the PERT formulas). This paper shows that some basic underlying assumptions behind most of these formulas are inappropriate; a more appropriate framework
Large deviations of the winsorized mean of an exponential distribution
โ Scribed by Stephen A. Book
- Publisher
- John Wiley and Sons
- Year
- 1974
- Tongue
- English
- Weight
- 291 KB
- Volume
- 21
- Category
- Article
- ISSN
- 0894-069X
No coin nor oath required. For personal study only.
โฆ Synopsis
Abstract
An asymptotic representation for large deviation probabilities of the Winsorized mean of a sequence of independent, identically distributed exponential random variables is derived. The Winsorized mean, a linear combination of exponential order statistics, is first transformed into a weighted sum of exponential random variables, and then a large deviation theorem for weighted sums can be applied. The representation obtained is then compared with results already known for the mean and the median, the two extreme cases of the Winsorized mean.
๐ SIMILAR VOLUMES
In this note we outlined the procedure to obtain upper tail percentage points of the distribution of the average ofk independent sample correlation coefficient (r) each of which is b e d on n pairs of observations. Null distribution of i demonstrates its application to medical research.