Multivariate Statistical Analysis And Combination Forecasting Method For Quality Evaluation

It’s well known that multivariate statistical analysis is widely used in management science,social science, life sciences, etc.. In the aspect of forecasting, multivariate statistical analysis and combination forecasting methods are more common. Based on multivariate statistical analysis and combination forecasting methods, wine quality will be evaluated in this paper. The main contents are as follows:1. This chapter is main about multivariate statistical methods, including ordinary least squares regression(OLSR), principal component regression(PCR), partial least squares regression(PLSR), and modified partial least squares regression(MPLSR). Then,this chapter introduces multivariate statistical combination forecasting methods, including optimal combination forecasting methods and sub-optimal combination forecasting methods. The optimal combination forecasting models are built based on the criteria of minimizing error sum of squares(MESS), minimizing the sum of absolute error(MSAE) and minimizing the maximum absolute error(MMAE). Moreover, several methods such as arithmetic average value method and the reciprocal prediction error sum of squares method are employed by the sub-optimal combination forecasting models.2. This chapter evaluates and predicts the wine quality by multivariate statistical methods and multivariate statistical combination forecasting methods based on actual data sets in winemaking process. Finally, by comparing the predicted results, we get the best model.