Research Article

An Empirical Likelihood Ratio-Based Omnibus Test for Normality with an Adjustment for Symmetric Alternatives

Table 1

Monte Carlo experiments to establish the values of at for the proposed CUSUM-type (CS) and Shiryaev–Roberts (SR) test statistics.

Distribution
CSSRCSSRCSSRCSSRCSSRCSSRCSSR

Symmetric short-tailed alternative distributions
Beta (2, 2)200.04860.03780.10720.11160.10200.10280.09280.08200.08840.10320.08380.08880.06340.0886
500.15280.10140.29040.29300.23660.26940.25360.22740.21760.29180.23720.26380.18220.2424
1000.35380.24800.56760.60620.51960.55120.54300.53840.53500.61940.51660.56880.41720.4968
Uniform (0, 1)200.15840.10560.29980.30300.22560.24020.25560.25580.21340.26280.21360.24040.13680.2132
500.52140.38820.79240.78740.67680.65900.75360.74820.69120.78040.69840.75600.54520.6714
1000.87500.80660.98820.98720.95960.95760.98380.98060.97360.98900.97380.98480.95320.9692
Trunc-norm (−1, 1)200.08980.06180.18780.19200.15120.16940.15600.15340.14460.17040.14040.14640.10040.1386
500.31400.20000.56040.56300.46240.48020.51440.50140.44900.56120.44660.49840.33840.4600
1000.66360.54120.88660.90940.82900.83980.86980.87060.85700.90380.85040.88660.77580.8358
Tukey (0, 1, 0.75, 0.75)200.11600.07260.22940.23320.17140.19040.18780.18820.16520.21180.17240.17660.10620.1640
500.38160.26400.64480.64580.52320.54420.62020.60020.53300.62060.55600.58800.40620.5334
1000.75980.65860.94280.94600.90140.89480.93620.92480.91320.95500.92280.93820.85680.8984
Symmetric long-tailed alternative distributions
T (4)200.23960.25300.22080.20600.18280.17820.22860.23420.19900.20420.19680.22540.17980.1850
500.44840.42120.47700.46000.40840.42200.44260.48100.44400.45960.43740.46840.41420.4302
1000.66300.65200.72640.70940.68960.69500.70620.71460.70180.71420.70120.72260.65560.6650
Logistic (0, 1)200.11920.11960.10760.10140.09020.08460.10500.11780.10380.09680.10020.10800.08880.0908
500.18400.17940.19900.19400.17080.15900.18900.19040.17360.18840.17900.19760.15460.1640
1000.27960.24900.30820.32160.27160.27660.30700.32540.29760.31040.29020.32040.24480.2712
Double-exp (0, 1)200.23960.27080.25980.22700.17720.17920.23640.26600.21160.20880.21960.23820.17320.1738
500.50320.47700.54460.52520.44100.44220.53220.54500.49260.48700.49800.52220.44120.4524
1000.76260.73340.81020.81580.74120.74700.80920.82140.79180.80580.78460.81300.73600.7328
Cauchy (0, 1)200.85880.85540.84080.83180.77420.75800.84640.86780.81380.81400.82260.84200.78200.7744
500.99400.99500.99580.99700.99440.99060.99560.99560.99460.99680.99380.99520.99360.9938
Asymmetric alternative distributions
Gamma (2, 1)500.86140.83060.56360.49000.28200.22400.81720.77820.52820.42300.77940.72540.56820.4202
1000.99500.99100.82660.76860.41820.33560.98900.98180.83720.72160.98540.97360.85360.7208
Weibull (2, 1)500.27940.28920.16800.13020.13540.11280.25680.23400.15900.13380.23580.21600.20700.1556
1000.58040.55060.22940.18700.17860.12980.50480.44960.27500.19500.46620.40620.32600.2106
SN (0, 1, 5)500.52680.50640.26740.20100.16880.14280.47720.43320.27500.20400.42620.38640.32560.2254
1000.87060.84020.42900.33920.21920.16460.84300.77320.47540.31660.80620.74400.54180.3506

Bold represents the most superior CS-type statistic and italicized represents the most superior SR statistic.