| dc.contributor |
Fakultas Matematika dan Ilmu Pengetahuan Alam |
|
| dc.creator |
Khairunnisa, Hilda |
|
| dc.creator |
Sunendiari, Siti |
|
| dc.date |
2019-01-25 |
|
| dc.date.accessioned |
2019-09-11T01:59:26Z |
|
| dc.date.available |
2019-09-11T01:59:26Z |
|
| dc.identifier |
http://karyailmiah.unisba.ac.id/index.php/statistika/article/view/14566 |
|
| dc.identifier.uri |
http://hdl.handle.net/123456789/22035 |
|
| dc.description |
Abstract. Forecasting is an estimate of something that hasn't happened yet. There are two methods that are usually used to forecast data, namely regression analysis and time series methods. One classic time series method is the Autoregressive Integrated Moving Average (ARIMA). The data analysis technique uses the ARIMA method because it is a technique for finding the most suitable pattern from a group of data. As time goes by, time series data forecasting methods are increasingly developing, one of which is the fuzzy time series method. In forecasting calculations using fuzzy time series, the length of the interval has been determined at the beginning of the calculation process and the length of the interval is made the same as the static interval. The disadvantages of using static intervals include historical data grouped into intervals roughly which results in forecasting being poor. Determination of interval length is very influential in the formation of fuzzy relationships so that it is made long every interval that is not static. Non-static interval is the determination of the length of each interval through a number of clustering algorithm processes, namely automatic clustering. In this study, fuzzy time series forecasting methods are applied, namely automatic clustering and fuzzy logical relationships and the ARIMA method and also comparing the accuracy of the two methods. The material to be used in this study is secondary data on the number of registrants and the number of new students registering at Bandung Islamic University from 2006 to 2018. From the analysis of data processing that has been done, the results of forecasting the number of new student applicants in 2019 using the automatic method are obtained. clustering and fuzzy logical relationships as many as 13,234 while the ARIMA method is 12,096 people and in the data of new students registering in 2019 using the method of automatic clustering and fuzzy logical relationships as many as 3,293 people while the ARIMA method is 3,148 people. Judging from the MAPE value, it was concluded that the automatic method of clustering and fuzzy logical relationships is an accurate method because it produces the smallest level of failure.Keywords: ARIMA, Fuzzy time series, fuzzy logical relationships, automatic clustering techniquesAbstrak. Peramalan (forecasting) adalah perkiraan sesuatu yang belum terjadi. Ada dua metode yang biasanya digunakan untuk meramalkan suatu data yaitu analisis regresi dan metode runtun waktu (time series). Salah satu metode time series klasik yaitu Autoregressive Integrated Moving Average (ARIMA). Teknik analisis data menggunakan metode ARIMA dilakukan karena merupakan teknik untuk mencari pola paling cocok dari sekelompok data. Seiring dengan berjalannnya waktu, metode peramalan data time series semakin mengalami perkembangan, salah satunya adalah metode fuzzy time series. Dalam perhitungan peramalan menggunakan fuzzy time series, panjang interval telah ditentukan di awal proses perhitungan dan panjang interval dibuat sama yang disebut dengan interval statis. Kekurangan menggunakan interval statis diantaranya data historis dikelompokan ke dalam interval secara kasar yang menghasilkan peramalan menjadi kurang baik. Penentuan panjang interval sangat berpengaruh dalam pembentukan fuzzy relationships sehingga dibuat panjang setiap interval yang tidak statis. Interval tidak statis adalah penentuan panjang setiap intervalnya melalui beberapa proses algoritma clustering yaitu dengan automatic clustering. Dalam penelitian ini, diterapkan metode peramalan fuzzy time series yaitu automatic clustering and fuzzy logical relationships dan metode ARIMA dan juga melakukan perbandingan tingkat akurasi dari kedua metode tersebut. Bahan yang akan digunakan dalam penelitian ini adalah data sekunder jumlah pendaftar dan jumlah mahasiswa baru yang melakukan registrasi di Universitas Islam Bandung dari tahun 2006 sampai 2018. Dari analisis pengolahan data yang telah dilakukan, diperoleh hasil peramalan jumlah pendaftar mahasiswa baru tahun 2019 dengan menggunakan metode automatic clustering and fuzzy logical relationships sebanyak 13.234 sedangkan metode ARIMA sebanyak 12.096 orang dan pada data mahasiswa baru yang melakukan registrasi tahun 2019 dengan menggunakan metode automatic clustering and fuzzy logical relationships sebanyak 3.293 orang sedangkan metode ARIMA sebanyak 3.148 orangDilihat dari nilai MAPE, disimpulkan bahwa metode automatic clustering and fuzzy logical relationships merupakan metode yang akurat karena menghasilkan tingkat kealahan yang paling kecil.Kata kunci : ARIMA, Fuzzy time series, fuzzy logical relationships, automatic clustering techniques |
|
| dc.format |
application/pdf |
|
| dc.language |
eng |
|
| dc.publisher |
Universitas islam Bandung |
|
| dc.relation |
http://karyailmiah.unisba.ac.id/index.php/statistika/article/view/14566/pdf |
|
| dc.rights |
Copyright (c) 2019 Prosiding Statistika |
|
| dc.source |
Prosiding Statistika; Vol 5, No 1, Prosiding Statistika (Februari, 2019); 40-46 |
|
| dc.source |
Prosiding Statistika; Vol 5, No 1, Prosiding Statistika (Februari, 2019); 40-46 |
|
| dc.source |
2460-6456 |
|
| dc.subject |
Statistika |
|
| dc.subject |
ARIMA, Fuzzy time series, fuzzy logical relationships, automatic clustering techniques |
|
| dc.title |
Perbandingan Metode Automatic Clustering and Fuzzy Logical Relationships dan Arima pada Peramalan Jumlah Pendaftar dan Jumlah Mahasiswa Baru yang Melakukan Registrasi di Universitas Islam Bandung |
|
| dc.type |
info:eu-repo/semantics/article |
|
| dc.type |
info:eu-repo/semantics/publishedVersion |
|
| dc.type |
Peer-reviewed Article |
|
| dc.type |
kuantitatif |
|