Desiccated coconut is one of the leading export products of Indonesia with increasing demand. The quality of desiccated coconut is determined by chemical content of lipids and free fatty acids (FFA). The measurement of the chemical content is mostly performed by chemical methods that leave chemical residues and caused health problems. The aim of this research was to use FT-NIR spectroscopy as a rapid method to estimate chemical content of lipids and FFA in desiccated coconut. The FT-NIR spectrometer with a 1000-2500 nm wavelength range was used to measure spectra by reflectance mode from 48 samples. The reference data was performed using the chemical method. The spectra data were pre-treated using SNV, MSC, second derivative, and normalization. The partial least square regression (PLSR) was used to determine the best calibration model to predict chemical content. The best calibration model for predicting chemical content were obtained using original spectra with 16 PLS factors with the accuracy indicators of r = 0.95, SEC = 0.05 % SEP = 0.55 %, CV = 0.85 %, RPD = 3.21 and consistency 82.81 for lipids content, and r= 0.90, SEC = 0.01 %, SEP = 0.01 %, CV= 5.89 %, RPD = 2.09 and consistency of 93.18 for free fatty acids content.