FUNCTION covariance (x , y ) { xmean := avg (x) ; ymean := avg (y) ; avg (( x - xmean ) * (y - ymean )) } FUNCTION sd ( x) { sqrt ( covariance (x , x) ) } FUNCTION pairCorr (x , y ) { covariance (x , y ) / ( sd (x) * sd (y )) } CREATE TABLE test1(a INT, b INT, c INT, d INT) LOAD DATA INFILE "data/test.csv" INTO TABLE test1 FIELDS TERMINATED BY "," SELECT pairCorr(c, b) * d, sum(a), b FROM test1 group by c,b,d order by b ASC