Evaluation of single missing value imputation approaches for incomplete air pollution data in Malaysia / Wan Suhailah Wan Mohamed Fauzi, Zuraira Libasin and Ahmad Zia ul-Saufie
This research is mainly focused on environmental scope, which is air pollution. İt is about evaluation of single missing value imputation approaches for incomplete air pollution data in Malaysia. Single missing value imputation means the replacement of blank space in monitoring dataset from chosen D...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/82889/1/82889.pdf https://ir.uitm.edu.my/id/eprint/82889/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Teknologi Mara |
| Language: | English |
| Summary: | This research is mainly focused on environmental scope, which is air pollution. İt is about evaluation of single missing value imputation approaches for incomplete air pollution data in Malaysia. Single missing value imputation means the replacement of blank space in monitoring dataset from chosen DOE monitoring station with calculated value from the best method for long gap hours. The variable that mainly being monitor is PM10. This variable is the primary source of air pollution release from industrial and transporation of everyday activities. Single imputation method focused in this research is mean imputation method. Furthermore, this methd will be tested on the dataset from Tanjung Malim monitoring station by fitting with many performance indicators such as MAE, RSME, R2, PA and IA. The result will be compared with previous study whether it is the best used for long gap hour data. Four stages need to be followed in order to complete this research. The steps are data acquisitions, characteristic analyzing of missing value, single imputation approach and lastly, verification of approach and suggestion of the best method. The four existing imputation method for missing data implemented in this research are series mean method, mean of nearby points, linear trend and linear interpolation. The finding from this research shows that interpolation method is the best method to be applied for particulate matter missing data replacement with least mean absolute error and the better in performance accuracy. |
|---|