Skip to main content

Table 5 Monte Carlo RMSEs and bias values of the entropy estimators VE ( m,n ) and AE ( m,n ) for the standard normal distribution and H(f) = 1.419 using DRSS

From: Goodness-of-fit testing for the inverse Gaussian distribution based on new entropy estimation using ranked set sampling and double ranked set sampling

n

m

VE (m,n)

AE (m,n)

  

Bias

RMSE

Bias

RMSE

10

1

−0.415021

0.472162

−0.352434

0.416211

 

2

−0.373395

0.412666

−0.262149

0.316029

 

3

−0.427401

0.459119

−0.254450

0.303820

 

4

−0.492911

0.518275

−0.264683

0.310442

 

5

−0.554351

0.577281

−0.267798

0.312339

20

1

−0.350703

0.383160

−0.323780

0.359592

 

2

−0.245907

0.277809

−0.190733

0.231106

 

3

−0.246496

0.276941

−0.158832

0.201924

 

4

−0.262789

0.290545

−0.148107

0.194728

 

5

−0.291340

0.317967

−0.145734

0.191755

 

6

−0.316105

0.341597

−0.147800

0.195946

 

7

−0.349246

0.373132

−0.150312

0.199934

 

8

−0.384526

0.406764

−0.152801

0.203493

 

9

−0.416151

0.436696

−0.156902

0.205954

 

10

−0.445901

0.465518

0.159050

0.207883

30

1

−0.321940

0.345223

−0.307781

0.332609

 

2

−0.206709

0.231560

−0.169564

0.198438

 

3

−0.187163

0.212774

−0.129694

0.163913

 

4

−0.190073

0.215577

−0.114103

0.152713

 

5

−0.199843

0.224569

−0.103570

0.145964

 

6

−0.214636

0.239021

−0.100510

0.146417

 

7

−0.231613

0.255278

−0.095517

0.143483

 

8

−0.247340

0.271084

−0.094560

0.145579

 

9

−0.268298

0.291044

−0.091548

0.145394

 

10

−0.286538

0.308661

−0.094236

0.149024

 

11

−0.305310

0.326485

−0.093843

0.150300

 

12

−0.324892

0.346062

−0.096171

0.152896

 

13

−0.343097

0.363236

−0.096892

0.153854

 

14

−0.369990

0.388586

−0.100541

0.155029

 

15

−0.387740

0.406081

−0.101202

0.156143

  1. Critical points at significance level 0.05 of the test statistic are given in Table 6. The optimal choice of the window size for a given sample size in the estimation of entropy using spacing's is still open problem for testing goodness-of-fit. The bold fonts in Table 6 are the largest critical values based on SRS, RSS and DRSS. For the suggested test, the optimal window size values are summarized in Table 7.