SSE/EFI Working Paper Series in Economics and Finance
No 548:
Estimating confidence regions over bounded domains
Bruno Eklund ()
Abstract: Estimating a density function over a bounded domain can be
very complicated and resulting in an unsatisfactory or unrealistic density
estimate. In many cases a one-to-one transformation can be applied to the
considered data set, but there are also situations where such a unique
transformation may not exist. This paper proposes a method to estimate
confidence regions over bounded domains when a one-to-one transformation
either does not exist or its existence is difficult to verify. By taking
into account parameter restrictions of a underlying model, a nonlinear grid
can be constructed, over which the density function can be estimated. The
method is illustrated by applying it to the kurtosis/first-order
autocorrelation of squared observations of the GARCH(1,1) model.
Keywords: Kernel estimation; nonlinear grid; GARCH model; highest density region; (follow links to similar papers)
JEL-Codes: C14; C22; (follow links to similar papers)
12 pages, November 28, 2003
Download Statistics
- This paper is published as:
-
Eklund, Bruno, (2005), 'Estimating confidence regions over bounded domains', Computational Statistics and Data Analysis, Vol. 49, pages 349-360
Questions (including download problems) about the papers in this series should be directed to Helena Lundin ()
Report other problems with accessing this service to Sune Karlsson ()
or Helena Lundin ().
Programing by
Design by Joachim Ekebom
