Skip to main content

Table 8 Empirical power estimates for three situations when h2 is 0.8. The empirical power estimates for three situations were calculated with 1000 replicates at the 0.01 significance levels. Phenotypes were generated under the assumption that the prevalence was assumed to 0.2. Prevalence was set to be 0.1, 0.2 or 0.3 to calculate the proposed statistics. The relative phenotypic variance attributable to the main disease gene was assumed to be 0.005

From: Adjusting heterogeneous ascertainment bias for genetic association analysis with extended families

n missing

n proband

Statistic

Prevalence to be set for statistics

0.1

0.2

0.3

0

1

FQLS 1

0.486

0.479

0.464

FQLS 2

0.572

0.539

0.525

2

FQLS 1

0.562

0.557

0.532

FQLS 2

0.606

0.621

0.624

3

FQLS 1

0.632

0.642

0.634

FQLS 2

0.644

0.647

0.654

1

1

FQLS 1

0.432

0.434

0.418

FQLS 2

0.468

0.469

0.462

2

FQLS 1

0.522

0.530

0.519

FQLS 2

0.547

0.554

0.559

3

FQLS 1

0.580

0.584

0.576

FQLS 2

0.575

0.584

0.585

2

1

FQLS 1

0.415

0.412

0.399

FQLS 2

0.428

0.420

0.412

2

FQLS 1

0.482

0.487

0.468

FQLS 2

0.483

0.495

0.492

3

FQLS 1

0.487

0.497

0.490

FQLS 2

0.500

0.515

0.511

  1. The bold text indicates the reference for the proposed statistics to compare with the misspecified prevalence