This is the European American single nucleotide polymorphisms (SNPs) data set Price, et al (2006), 
which consists of 488 European American samples. This data set was used in Section 7.2.5.
The response is the height phenotype (0/1, binary variable) of these European American samples.
It is of interest to find variables that are associated with this phenotype among a set of 277 SNPs. 
The genotype for each SNP is a categorical variable, coded as 0/1/2. As in Price et al (2006), 
the outlier individuals are removed. This leads to a total of 361 observations. However, for each 
observation, approximately 2% of SNPs are missing on average.  We imputed all the missing values 
using the R package MissForest available in  CRAN. This package uses a random forest trained 
based on the observed entries  to predict those missing values.

Response: height.pick300.pheno

Covariates are SNPs in genotype.pick300.dat