PERBANDINGAN METODE DEKOMPOSISI KLASIK DENGAN METODE PEMULUSAN EKSPONENSIAL HOLT-WINTER DALAM MERAMALKAN TINGKAT PENCEMARAN UDARA DI KOTA BANDUNG PERIODE 2003-2012

Hapsari, Vanissa (2013) PERBANDINGAN METODE DEKOMPOSISI KLASIK DENGAN METODE PEMULUSAN EKSPONENSIAL HOLT-WINTER DALAM MERAMALKAN TINGKAT PENCEMARAN UDARA DI KOTA BANDUNG PERIODE 2003-2012. Other thesis, Universitas Pendidikan Indonesia.

[img]
Preview
Text
S_MTK_0905655_Title.pdf

Download (765kB) | Preview
[img]
Preview
Text
S_MTK_0905655_Abstract.pdf

Download (430kB) | Preview
[img]
Preview
Text
S_MTK_0905655_Table_Of_Content.pdf

Download (313kB) | Preview
[img]
Preview
Text
S_MTK_0905655_Chapter1.pdf

Download (459kB) | Preview
[img] Text
S_MTK_0905655_Chapter2.pdf
Restricted to Repository staff only

Download (622kB)
[img]
Preview
Text
S_MTK_0905655_Chapter3.pdf

Download (557kB) | Preview
[img] Text
S_MTK_0905655_Chapter4.pdf
Restricted to Repository staff only

Download (655kB)
[img]
Preview
Text
S_MTK_0905655_Chapter5.pdf

Download (318kB) | Preview
[img]
Preview
Text
S_MTK_0905655_Bibliography.pdf

Download (201kB) | Preview
[img] Text
S_MTK_0905655_Appendix.pdf
Restricted to Repository staff only

Download (435kB)

Abstract

Data yang diamati berdasarkan urutan waktu merupakan data runtun waktu. Peramalan pada data runtun waktu dapat dilakukan untuk memprediksi sejumlah data di masa mendatang. Akan dilakukan peramalan tingkat pencemaran udara di kota Bandung menggunakan metode Dekomposisi Klasik dan metode Eksponensial Holt-Winter. Pengambilan sampel partikulat kasar berukuran 2,5μ-10μ secara berkala dilakukan untuk mengetahui tingkat pencemaran udara di suatu tempat. Metode Dekomposisi merupakan suatu metode yang memecah data berdasarkan komponen musiman, trend dan siklus, dengan persamaan F_t=S_t×T_t×C_t. Pada metode Holt-Winter memiliki 3 parameter pembobotan untuk pemulusan unsur musiman, trend dan keseluruhan data, diperoleh persamaan F_(120+m)=(23264,79+(-75,7953)m) S_(108+m). Untuk mengetahui metode mana yang paling baik untuk meramalkan suatu data, digunakan alat ukur diantaranya Mean Squared Deviation (MSD), Mean Absolute Deviation (MAD), dan Mean Absolute Percentage Error (MAPE) dan metode yang memiliki tingkat kesalahan terkecil yang dipilih. Berdasarkan hasil perhitungan didapatkan untuk metode Dekomposisi memiliki nilai MSD = 51831705,862, MAD = 5487,835 dan MAPE = 31,904 sementara untuk metode Holt-Winter memiliki nilai MSD = 34749831, MAD = 4170,602 dan MAPE = 28,411. Observed data based upon time is called time series data. Forecasting time series data is done to predict data in a future as preventive action. Clasic decomposition method and Holt-Winter’s method will be used to forecast air pollution level in Bandung. Coarse particulat (2,5μ-10μ) sampling conducted regularly to know air pollution level in some place. Decomposition method try to identify seasonality, trend and cycle component which assumes that the data has the form F_t=S_t×T_t×C_t. Holt-Winter’s method involves three smoothing parameters to smooth the data, the trend and the seasonal index, in this case the form is F_(120+m)=(23264,79+(-75,7953)m) S_(108+m). Several measurement tools are used which is Mean Squared Deviation (MSD), Mean Absolute Deviation (MAD), and Mean Absolute Percentage Error (MAPE) A method that has the least number of measurement will be the best method to predict the data. Based on the calculation, Decomposition method have MSD = 51831705,862, MAD = 5487,835 and MAPE = 31,904 while Holt-Winter’s method have MSD = 34749831, MAD = 4170,602 and MAPE = 28,411.

Item Type: Thesis (Other)
Subjects: Universitas Pendidikan Indonesia > Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Jurusan Pendidikan Matematika > Program Studi Pendidikan Matematika
Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Jurusan Pendidikan Matematika > Program Studi Pendidikan Matematika
Depositing User: DAM STAF Editor
Date Deposited: 06 Nov 2013 02:41
Last Modified: 06 Nov 2013 02:41
URI: http://repository.upi.edu/id/eprint/2869

Actions (login required)

View Item View Item