Wikipedia 10K Redux by Reagle from Starling archive. Bugs abound!!!

<-- Previous | Newer --> | Current: 983447894 Dick Beldin at Thu, 01 Mar 2001 11:58:14 +0000.

Estimating_Parameters

back to [[Statistical Theory]] -- [[Applied Statistics]]

In general, we assume that we know everything about an situation we are observing except for the value of a certain [[numerical parameter]]. We use the data to get information about that parameter. The most common case is that of the [[location parameter]] of a [[Normal Distribution]] which we estimate by the [[Arithmetic Mean]] of the observations. 

When we just quote the estimated value, we have a [[point estimate]]. More often, we wish to find an interval with a high probability of covering the value of the parameter. This is a [[confidence interval]], a specific form of an [[interval estimate]]. 
----
[[Dick Beldin]]