A method for determination of palmitic acid in edible oils by the near-infrared spectroscopy was addressed in this paper. 56 samples were collected in the experiment. In terms of concentration content gradient method, 44 samples were selected for modeling set and 12 for testing set. This paper described the utilization of PLS for establishing a quantitative analysis model for predicting the content of palmitic acid in edible oils by near-infrared spectroscopy. The result shows that the model has a high accuracy for predicting the palmitic acid content with vector normalization and first-derivative preprocessing spectra with its best main factorial number of 8.The determination coefficients (R2), root mean square error of cross validation (RMSECV), root mean square error of prediction (RMSEP) and average bias are 99.59, 0.162, 0.306 and 0.148, respectively. The model used in the paper can be adopted for the measurement of palmitic acid in edible oils accurately.