Loading...
Please wait, while we are loading the content...
Similar Documents
1 Goodness-ofFit Test for Normality Using the Minimum Hellinger Distance
| Content Provider | Semantic Scholar |
|---|---|
| Copyright Year | 2015 |
| Abstract | Statistical data analysis techniques rest on assumptions about the form of the data. Some techniques are sensitive to small deviations from the assumptions while others are more robust. Unfortunately, robust techniques are not usually optimal when the data follow the true model. However, one robust estimator (beran1977a) (beran1977b) based on the minimum Hellinger distance between a parametric family of densities and a nonparametric density estimator (called the mhde) is also asymptotically efficient (stather1981). The minimized Hellinger distance can also be used in a goodness-of-fit test for the parametric family (beran1977a). Empirically based critical values for a goodness-of-fit test for normality based on the Hellinger distance were developed (eslinger1991) using a limited number of simulations and they are not readily accessible to the general community of practitioners. This paper updates and expands the critical values for the minimum Hellinger distance goodness-of-fit test for normality and introduces the mhde R package (mhde). The package determines the mean and standard deviation of the normal distribution that minimizes the Hellinger distance and calculates the p-value for a goodness-of-fit test. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://cran.r-project.org/web/packages/mhde/vignettes/mhde_article.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
