The imputation of missing tides data in Malaysian tourism areas using basic statistical methods

This study analyses the imputation of missing tide data in Malaysian tourism areas using basic statistical methods. It aims to determine the most appropriate method among five basic statistical methods for the imputation of missing tide data in three Malaysian tourist areas, namely Kota Kinabalu, Penang and Langkawi Island. These methods are Top Bottom Mean, 6-Hour Mean, 12-Hour Mean, Daily Mean, and Linear Interpolation. The data were recorded hourly for 14 days, which is equivalent to 336 hours, in 2019. The data are complete and continuous. The percentage of data discarded in this study is 10%, 20%, 30%, 40%, and 50%. The performance indices used to evaluate the methods are Correlation Coefficient (CC), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). Overall, the best basic statistical method to impute missing tide data is Linear Interpolation, and it is hoped that this study can help the Department of Survey and Mapping Malaysia (JUPEM) in imputing the missing tide data.