Research Article
An Empirical Likelihood Ratio-Based Omnibus Test for Normality with an Adjustment for Symmetric Alternatives
Table 3
Type I error rates along with the standard error for the proposed test statistics using
.
| Alpha | Simulated probabilities of the type I error along with standard error | Sample size | CUSUM-type test statistic | Shiryaev–Roberts test statistic |
| 0.01 | 20 | 0.009214 1.3509e − 04 | 0.009346 1.3606e − 04 | 30 | 0.010163 1.4181e − 04 | 0.010291 1.4269e − 04 | 50 | 0.010589 1.4469e − 04 | 0.009849 1.3963e − 04 | 80 | 0.010893 1.4675e − 04 | 0.009613 1.3796e − 04 | 100 | 0.009706 1.3862e − 04 | 0.009916 1.4003e − 04 | 0.05 | 20 | 0.050133 3.0859e − 04 | 0.049785 3.0758e − 04 | 30 | 0.049841 3.0774e − 04 | 0.050578 3.0989e − 04 | 50 | 0.052238 3.1464e − 04 | 0.049027 3.0534e − 04 | 80 | 0.051434 3.1235e − 04 | 0.050018 3.0826e − 04 | 100 | 0.048295 3.0317e − 04 | 0.051756 3.1327e − 04 | 0.10 | 20 | 0.102066 4.2812e − 04 | 0.100521 4.2523e − 04 | 30 | 0.100938 4.2601e − 04 | 0.098578 4.2156e − 04 | 50 | 0.101902 4.2781e − 04 | 0.099039 4.2242e − 04 | 80 | 0.101586 4.2722e − 04 | 0.099829 4.2392e − 04 | 100 | 0.095552 4.1573e − 04 | 0.101395 4.2686e − 04 |
|
|