Higher-order approximate confidence intervals
| Autor: | Silvia Ferrari, Francisco M.C. Medeiros, Eliane C. Pinheiro |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Statistics and Probability Monte Carlo method Asymptotic distribution Score Mathematics - Statistics Theory Statistics Theory (math.ST) Statistics - Applications Statistics - Computation 01 natural sciences 010104 statistics & probability Approximation error 0502 economics and business FOS: Mathematics Applied mathematics Order (group theory) Applications (stat.AP) 0101 mathematics Computation (stat.CO) 050205 econometrics Mathematics Applied Mathematics 05 social sciences Regression analysis INFERÊNCIA PARAMÉTRICA Confidence interval Statistics Probability and Uncertainty Quantile |
| Zdroj: | Repositório Institucional da USP (Biblioteca Digital da Produção Intelectual) Universidade de São Paulo (USP) instacron:USP |
| ISSN: | 0378-3758 |
| DOI: | 10.1016/j.jspi.2020.11.013 |
| Popis: | Standard confidence intervals employed in applied statistical analysis are usually based on asymptotic approximations. Such approximations can be considerably inaccurate in small and moderate sized samples. We derive accurate confidence intervals based on higher-order approximate quantiles of the score function. The coverage approximation error is $O(n^{-3/2})$ while the approximation error of confidence intervals based on the asymptotic normality of MLEs is $O(n^{-1/2})$. Monte Carlo simulations confirm the theoretical findings. An implementation for regression models and real data applications are provided. Comment: 25 pages, 9 figures |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect