Fig. 3From: Statistically efficient association analysis of quantitative traits with haplotypes and untyped SNPs in family studiesType I error and power of association tests at SNP 20, which has a causal additive effect on the phenotype, and SNP 21, which is null, at the 1% nominal significance level when there are 10% missing genotype data. For complete-data analysis, all subjects with missing data are removed. For full-data analysis, the missing genotypes are replaced by their true valuesBack to article page