Fig. 2

Predictor selection by the least absolute shrinkage and selection operator (LASSO). a Parameter (Lambda) selection by LASSO adopted tenfold cross-validation via minimum criteria. Dotted vertical lines were drawn at the optimal values by adopting the minimum criteria and the 1 standard error of the minimum criteria (the 1 − SE criteria). The Lambda value of 0.071, with log (Lambda), − 1.478 was chosen (1 − SE criteria) by tenfold cross-validation. b LASSO coefficient profile plot of 55 variables against the log (Lambda) sequence. Vertical line was drawn at optimal Lambda value with 4 nonzero coefficients by tenfold cross-validation