Method for multivariate analysis with small sample in aircraft cost estimation

Partial least-squares regression (PLSR) is applied to multivariate analysis and parametric modeling for estimating aircraft cost-estimating relationship. PLSR method is a statistical tool that has been specially designed to deal with multiple regression where the number of observations is limited, missing data are numerous, and the correlations between variables are high. It is a recent technique that generalizes and combines features from principal component analysis and multiple least-squares regression. Through finding out the outliers in observations, VIP analysis, and the explanatory variable selection, the good relationship for estimating aircraft cost can be derived by PLSR. The application of PLSR to aircraft cost estimation can play a very important role in the cost data analysis.