RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY
Pangkep Regency is one of the regions in South Sulawesi that is the center of national salt production. Salt production in the area is still dependent on sea water evaporation so that rainfall is one of the determining factors for the success of salt productivity. Statistical downscaling is an accur...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universitas Pattimura
2024-03-01
|
| Series: | Barekeng |
| Subjects: | |
| Online Access: | https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/10814 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849407887678373888 |
|---|---|
| author | Sitti Sahriman Eunike Laurine Randa Sitti Aisyah Surianda M. Zaky Gozhi Hisyam Muh. Ikbal Taufik Guntur Dwi Putra |
| author_facet | Sitti Sahriman Eunike Laurine Randa Sitti Aisyah Surianda M. Zaky Gozhi Hisyam Muh. Ikbal Taufik Guntur Dwi Putra |
| author_sort | Sitti Sahriman |
| collection | DOAJ |
| description | Pangkep Regency is one of the regions in South Sulawesi that is the center of national salt production. Salt production in the area is still dependent on sea water evaporation so that rainfall is one of the determining factors for the success of salt productivity. Statistical downscaling is an accurate method for rainfall forecasting by linking the local scale rainfall in Pangkep Regency (response variable) with the global scale of the global circulation model/GCM output (predictor variable). However, the GCM output rainfall has a large dimension, which is an 8×8 grid (64 predictor variables), causing multicollinearity. The linearized ridge regression (LRR) method is used to overcome this problem. This method combines the performance of generalized ridge regression and Liu-type methods to reduce multicollinearity. In addition, dummy variables based on the K-means clustering technique were added to the model to overcome heteroscedasticity. The purpose of this study is to obtain the results of rainfall forecasting in Pangkep Regency using the LRR method in the statistical downscaling model. The model generated from the LRR method with dummy variables is better at explaining the variability of rainfall in Pangkep Regency. The value is higher (72%) than without dummy variables (57%). The addition of dummy variables in the LLR model has better accuracy in forecasting rainfall. The actual rainfall correlation of Pangkep Regency with has the largest correlation (0.76) with the smallest mean absolute percentage error value (0.49). The results obtained are that the months of May - November tend to have relatively low rainfall, so that salt farmers can produce salt with good quantity and quality. |
| format | Article |
| id | doaj-art-ba17c16977564cf0a2a013a26f90c8b5 |
| institution | Kabale University |
| issn | 1978-7227 2615-3017 |
| language | English |
| publishDate | 2024-03-01 |
| publisher | Universitas Pattimura |
| record_format | Article |
| series | Barekeng |
| spelling | doaj-art-ba17c16977564cf0a2a013a26f90c8b52025-08-20T03:35:54ZengUniversitas PattimuraBarekeng1978-72272615-30172024-03-011810483049210.30598/barekengvol18iss1pp0483-049210814RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMYSitti Sahriman0Eunike Laurine Randa1Sitti Aisyah Surianda2M. Zaky Gozhi Hisyam3Muh. Ikbal Taufik4Guntur Dwi Putra5Statistics Study Program, Department of Statistics, Faculty of Mathematics and Natural Sciences, Hasanuddin University, IndonesiaStatistics Study Program, Department of Statistics, Faculty of Mathematics and Natural Sciences, Hasanuddin University, IndonesiaStatistics Study Program, Department of Statistics, Faculty of Mathematics and Natural Sciences, Hasanuddin University, IndonesiaStatistics Study Program, Department of Statistics, Faculty of Mathematics and Natural Sciences, Hasanuddin University, IndonesiaStatistics Study Program, Department of Statistics, Faculty of Mathematics and Natural Sciences, Hasanuddin University, IndonesiaGeophysics Study Program, Department of Geophysics, Faculty of Mathematics and Natural Sciences, Hasanuddin University, IndonesiaPangkep Regency is one of the regions in South Sulawesi that is the center of national salt production. Salt production in the area is still dependent on sea water evaporation so that rainfall is one of the determining factors for the success of salt productivity. Statistical downscaling is an accurate method for rainfall forecasting by linking the local scale rainfall in Pangkep Regency (response variable) with the global scale of the global circulation model/GCM output (predictor variable). However, the GCM output rainfall has a large dimension, which is an 8×8 grid (64 predictor variables), causing multicollinearity. The linearized ridge regression (LRR) method is used to overcome this problem. This method combines the performance of generalized ridge regression and Liu-type methods to reduce multicollinearity. In addition, dummy variables based on the K-means clustering technique were added to the model to overcome heteroscedasticity. The purpose of this study is to obtain the results of rainfall forecasting in Pangkep Regency using the LRR method in the statistical downscaling model. The model generated from the LRR method with dummy variables is better at explaining the variability of rainfall in Pangkep Regency. The value is higher (72%) than without dummy variables (57%). The addition of dummy variables in the LLR model has better accuracy in forecasting rainfall. The actual rainfall correlation of Pangkep Regency with has the largest correlation (0.76) with the smallest mean absolute percentage error value (0.49). The results obtained are that the months of May - November tend to have relatively low rainfall, so that salt farmers can produce salt with good quantity and quality.https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/10814dummyglobal circulation modellinearized ridge regressionmulticollinearityrainfallstatistical downscaling |
| spellingShingle | Sitti Sahriman Eunike Laurine Randa Sitti Aisyah Surianda M. Zaky Gozhi Hisyam Muh. Ikbal Taufik Guntur Dwi Putra RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY Barekeng dummy global circulation model linearized ridge regression multicollinearity rainfall statistical downscaling |
| title | RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY |
| title_full | RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY |
| title_fullStr | RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY |
| title_full_unstemmed | RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY |
| title_short | RAINFALL FORECASTING OF SALT PRODUCING AREAS IN PANGKEP REGENCY USING STATISTICAL DOWNSCALING MODEL WITH LINEARIZED RIDGE REGRESSION DUMMY |
| title_sort | rainfall forecasting of salt producing areas in pangkep regency using statistical downscaling model with linearized ridge regression dummy |
| topic | dummy global circulation model linearized ridge regression multicollinearity rainfall statistical downscaling |
| url | https://ojs3.unpatti.ac.id/index.php/barekeng/article/view/10814 |
| work_keys_str_mv | AT sittisahriman rainfallforecastingofsaltproducingareasinpangkepregencyusingstatisticaldownscalingmodelwithlinearizedridgeregressiondummy AT eunikelaurineranda rainfallforecastingofsaltproducingareasinpangkepregencyusingstatisticaldownscalingmodelwithlinearizedridgeregressiondummy AT sittiaisyahsurianda rainfallforecastingofsaltproducingareasinpangkepregencyusingstatisticaldownscalingmodelwithlinearizedridgeregressiondummy AT mzakygozhihisyam rainfallforecastingofsaltproducingareasinpangkepregencyusingstatisticaldownscalingmodelwithlinearizedridgeregressiondummy AT muhikbaltaufik rainfallforecastingofsaltproducingareasinpangkepregencyusingstatisticaldownscalingmodelwithlinearizedridgeregressiondummy AT gunturdwiputra rainfallforecastingofsaltproducingareasinpangkepregencyusingstatisticaldownscalingmodelwithlinearizedridgeregressiondummy |