Quantitative structure-retention relationship study for prediction of retention indices of essential oils ofNeolitsea fischeri Gamble by multiple linear regression

Quantitative structure-retention relationship (qsrr) study for prediction ofgc retention indices ofessential oil compounds were developed by application ofthe structural descriptors and multiple linearregression (mlr) method. theoretical molecular descriptors selected by a stepwise-mlr method. a simplemodel with low standard errors and high correlation coefficients was selected. the proposed methodologywas validated using leave-one-out and leave-group-out cross validation using division ofthe available dataset into training and test sets. the results illustrated that the linear techniques such as mlr combined with asuccessful variable selection procedure are capable to generate an efficient qsrr modelfor predicting theretention indices ofdifferent compounds. this model, with high statistical significance (ifcalibration=0. 963,ifprediction=0. 943, rzloo=0.882, rzlgo=o. 936, rep(%)=2.928), could be used adequately for the predictionand description ofthe retention indices ofessential oil compounds