By analyzing alcohol extractive content, density, GC-MS and HPLC data of agarwood samples in different qualities, agarwood quality factors were determined. Then an agarwood grading model was established by combining multivariate analysis methods, such as principal component analysis（PCA）,linear discriminant analysis（LDA） and random forest. The results showed that the contents of alcohol extractive, chromones as well as sesquiterpene type and contents from the Sin Chew system agarwood in different quality were obviously distinct. Therefore, factors including the content of alcohol extractive, the total content of sesquiterpenoids, and the content of 2-（2-phenethyl） chromone could be used for quality classification of agarwood.