[Chemical structural characterization of some components from essential oil of Rosa banksiae for estimation and prediction of their retention time].

The molecular electronegativity-distance vector (MEDV) was used to describe the chemical structural characterization of 46 components of essential oils in the flower of Rosa banksiae. Various multiple linear regression (MLR) models were created with variable screening by the stepwise multiple regression technique and statistics. The QSRR models of 10 and 6 variables were built by MLR with the correlation coefficients (R) of molecular modeling being 0.906 and 0.903. Cross-validation of the models, which contain selected vectors were performed by leave-one -out procedure (LOO) and the satisfied results with correlation coefficients (Rcv) of 0.904 and 0.903, respectively. The results showed that the models constructed can provide estimation stability and favorable predictive ability.