A Study on Bayesian Principal Component Analysis for Addressing Missing Rainfall Data

This paper proposed the application of Bayesian Principal Component Analysis (BPCA) algorithm to address the issue of missing rainfall data in Kuching City. The experiment was conducted using six different combinations of rainfall data from different neighbouring rainfall stations at different missing data entries (1%, 5%, 10%, 15%, 20%, 25% and 30% of missing data entries). The performance of BPCA model in reconstructing the missing data was examined with respect to Bias (B-s), Efficiency (E) and Root Mean Square Error (RMSE). The reliability and robustness of BPCA was confirmed by comparing its performance with K-Nearest Neighbour (KNN) imputation model. The results support the addition of data from neighbouring rainfall stations to improve the imputation accuracy.