Missing data (Statistics)

This study focuses on developing optimization models to estimate missing precipitation data at twenty-two sites within Kentucky State. Various optimization formulations and regularization models are explored in this context. The performance of these models is evaluated using a range of performance measures and error metrics for handling missing records. The findings revealed that regularization models performed better than optimization models. This superiority is attributed to their ability to reduce model complexity while enhancing overall performance. The study underscores the significance of regularization techniques in improving the accuracy and efficiency of precipitation data estimation.

Missing rainfall records happens frequently in many areas, and making precipitation estimation has been a challenge due to the spatial-temporal variability of the parameter. Model tree (MT), regression tree (RT), and ensemble approach models were developed and evaluated for estimating missing precipitation values in this research study. The selection of stations using correlation coefficient and similar distribution, and variation of data used to build the model were applied in this study. Proposed models were developed and tested using daily rainfall data from 1971 to 2016 at twenty-two stations in Kentucky, U.S.A. The model results were analyzed and evaluated using error and performance measures. The results indicated that MT-based and ensemble models produce a better estimation of missing rainfall than regression trees. The MT-based model was able to estimate missing rainfall accurately without needing objective selection of stations and using minimal calibration data to build the model.