作者: D. J. Venzon , S. H. Moolgavkar
DOI: 10.2307/2347496
关键词:
摘要: SUMMARY The method of constructing confidence regions based on the generalised likelihood ratio statistic is well known for parameter vectors. A similar construction a interval single entry vector can be implemented by repeatedly maximising over other parameters. We present an algorithm finding these endpoints that requires less computation. It employs modified Newton-Raphson iteration to solve system equations defines endpoints.