Loading...
Please wait, while we are loading the content...
Random Walk Approximation of Conndence Intervals
| Content Provider | Semantic Scholar |
|---|---|
| Author | Murdoch, Duncan J. |
| Copyright Year | 1996 |
| Abstract | Construction of likelihood-based conndence intervals and regions is a computationally intensive task. In low dimensions (up to 2 or 3 parameters), an eeective method is to evaluate the likelihood over a dense grid. In higher dimensions proole-based methods (Bates and Watts, 1988) are eeective when it is practical to parametrize the model in terms of the quantities of interest, but it is diicult to construct conndence intervals for general functions of the parameters. In this paper we develop an algorithm for approximation of conndence intervals which is computationally eecient in models with up to about 10 parameters. The algorithm is based on a variation on Gibbs sampling (Gelfand and Smith, 1990) of a uniform distribution on a conndence region in the full parameter space and uses extrapolated QQ plots to adjust the borders of the resulting regions. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://fisher.stats.uwo.ca/faculty/murdoch/research/papers/walkconf.ps |
| Language | English |
| Access Restriction | Open |
| Subject Keyword | Algorithm Approximation Dimensions Extrapolation Gibbs sampling Population Parameter Quantity Sampling (signal processing) Tencent QQ Watt |
| Content Type | Text |
| Resource Type | Article |