Skip to main content

Table 7 Simulation of 1000 trials when the COI is L-G

From: Connecting a disconnected trial network with a new trial: optimizing the estimation of a comparative effect in a network meta-analysis

 

Best Connecting Trial

Direct Trial

n

Trial

Bias\((\hat{\mu }_{L,G})\)

RMSE\((\hat{\mu }_{L,G})\)

Bias\((\hat{\mu }_{L,G})\)

RMSE\((\hat{\mu }_{L,G})\)

6

DJ-G

0.5128

1.3339

-1.5278

2.8869

8

H-G

0.3076

1.1185

-1.4155

2.5219

10

H-G

0.2441

0.9768

-1.3064

2.3086

12

H-G

0.2437

0.9272

-1.2617

2.1308

14

H-G

0.1923

0.8742

-1.1754

1.9721

16

H-G

0.1592

0.8097

-1.1510

1.8554

18

H-G

0.1401

0.8225

-1.0223

1.5928

20

N-G

0.1063

0.7047

-1.0123

1.5995

Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us