Imputation of Missing Daily Rainfall Data; A Comparison Between Artificial Intelligence and Statistical Techniques

The acquisition of a comprehensive and extensive rainfall dataset is crucial in ensuring the effective completion of a hydrological study. This study examines different statistical and artificial intelligencebased techniques (AITs) for imputing missing daily rainfall data. This study evaluated daily rainfall data collected at twenty stations from northern Thailand. The evaluation of imputation methods was conducted through the mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (R2), and correlation (r).

The experimental findings revealed that the MLR and M5-MT techniques exhibited promising performance. Overall, For the MLR model, the average MAE was approximately 0.98, the average RMSE was around 4.52, and the average R2 was about 79.6the average MAE was approximately 0.91, the average RMSE was about 4.52, and the average R2 value was around 79.8recommended approach due to its ability to deliver good estimation results while offering a transparent mechanism and not necessitating prior knowledge for model creation. This study used Statistical techniques (STs), including arithmetic averaging (AA), multiple linear regression (MLR), normal-ratio (NR), nonlinear iterative partial least squares (NIPALS) algorithm, and linear interpolation were used. STs results were compared with AITs, including long-short-term-memory recurrent neural network (LSTMRNN), M5 model tree (M5-MT), multilayer perceptron neural networks (MLPNN), support vector regression with polynomial and radial basis function SVR-poly and SVR-RBF.