PERBANDINGAN METODE PERAMALAN MENGGUNAKAN SINGLE EXPONENTIAL SMOOTHING DAN RANDOM FOREST PADA DATA OUTLIER

Authors

  • Lukmanul Hakim Prodi Sains Data Universitas Insan Cita Indonesia
  • Asep Saefuddin IPB University
  • Ristu Haiban Hirzi Universitas Hamzanwadi
  • Agustifa Zea Tazliqoh Universitas Singaperbangsa Karawang

DOI:

https://doi.org/10.32493/sm.v4i2.27349

Keywords:

Exponential, Smoothing, Random Forest, MAPE, Forecasting

Abstract

ABSTRACT

 

The progress of a country is seen from various indicators and one of them is the welfare of its population. The most basic welfare of the population in an agrarian country like Indonesia can be seen from the welfare of its farmers. The indicator that is commonly used to measure the welfare of farmers is by using Farmer Exchange Rates (NTP). However, it is known that the exchange rate of farmers during the Covid 19 pandemic has experienced a very drastic decline. This is difficult for the government to make predictions. So a special method is needed in handling it. In this study, two methods were used, namely single exponential smoothing and random forest. From the research results, it was found that the MAPE value in single exponential smoothing was smaller when compared to the random forest. However, the fact is that the exchange rate of farmers every year always increases. Therefore it can be concluded that exponential smoothing is weak against outlier data.Keywords: Exponential ,Smoothing, Random Forest, MAPE, Forecasting

 

 

ABSTRAK

Maju atau tidaknya suatu negara dilihat dari berbagai indikator dan salah satunya yaitu kesejahteraan penduduknya. Kesejahteraan penduduk yang paling mendasar pada negara agraris seperti Indonesia dapat dilihat dari kesejahteraan petaninya. Indikator yang umum digunakan untuk mengukur kesejahteraaan petani yaitu dengan menggunakan Nilai Tukar Petani (NTP). Akan tetapi diketahui bahwa nilai tukar petani selama pandemi covid 19 melangalami penurunan yang sangat drastis. Hal ini sulit pagi pemerintah dalam melakukan prediksi. Sehingga di butuhkan metode khusus dalam penanganannya. Dalam penelitian ini menggunakan dua metode yaitu singgel exponential smoothing dan random forest. Dari hasil penelitian didapatkan hasil bahwa nilai MAPE pada single exponential smoothing lebih kecil jika dibandingkan dengan random forest. Akan tetapi faktanya nilai tukar petani setiap tahunnya selalu mengalami peningkatan. Oleh karena itu dapat disimpulkan bahwa exponential smoothing lemah terhadap data outlier.

 

Kata Kunci: Exponential ,Smoothing, Random Forest, MAPE, Peramalan

 

References

Abdurrahman R dan Hakim L. 2021. Analisa Nilai Tukar Petani Di Provinsi Riau. Jurnal Ilmiah Komputerisasi Akuntansi, Vol. 14, No. 2.

Aulia S.S, Rimbodo D.S, dan Wibowo M.G. 2021. Faktor-faktor yang Memengaruhi Nilai Tukar Petani (NTP) di Indonesia. Journal of Economics and Business Aseanomics (16).

Badan Pusat Statistik. 2013. Statistik Nilai Tukar Petani.

Breiman L, Cutler A. 2003. Manual on Setting Up, Using, and Understanding Random Forest V4.0.

Breiman L. 2001. Random Forests. Machine Learning 45:5-32

Ekaria dan Hasyyati A.N. 2014. Kajian Penghitungan Nilai Tukar Petani Tanaman Pangan (NTPP) di Jawa, Bali, dan Nusa Tenggara Tahun 2011- 2013. Jurnal Aplikasi Statistika dan Komputasi Statistik.

Hanzak T dan Cipra T. 2011. Exponential smoothing for time series with outliers.

Muhammad N.S dan Din A.M. 2017. Exponential Smoothing Techniques On Dailytemperature Level Data. ICOCI.

Nirmala A.R, Hanani N, dan Muhaimin A.W. 2016. Analisis Faktor Faktor yang Mempengaruhi Nilai Tukar Petani Tanaman Pangan di Kabupaten Jombang. Jurnal Habitat. Volume 27 no 2.

Olivia M dan Amelia. 2021. Metode Exponential Smoothing Untuk Forecasting Jumlah Penduduk Miskin Di Kota Langsa. Gamma-Pi: Jurnal Matematika dan Terapan. Volume 3 Nomor 1.

Raihan, Eff M.S dan Hendrawan A. 2016. Forcasting Model Exsponensial Smoothing Time Series Rata Rata Mechanical Availability Unit Off Highway Truck Cat 777d Caterpillar. Jurnal POROS TEKNIK. Volume 8, No. 1.

Riyandh M.I. 2015. Analisis Nilai Tukar Petani Komoditas Tanaman Pangan Di Sumatera Utara.

Sopiyan M, Fauziah, Wijaya Y.F. 2022. Fraud Detection Using Random Forest Classifier, Logistic Regression, and Gradient Boosting Classifier Algorithms on Credit Cards . JUITA: Jurnal Informatika. Vol 10. No 1.

Zhu T. 2020. Analysis on the Applicability of the Random Forest. Journal of Physics: Conference Series

Downloads

Published

2022-07-31

Issue

Vol. 4 No. 2 (2022)

Section

Articles

License

As an Author, you have the right to a variety of uses for your article, including institutions or companies. The author's rights might do without the need for special permission.

 

Authors who publish in the Jurnal Jurnal Statistika dan Matematika (Statmat) have broad rights to use their works for education and scientific purposes without permission, including:

  1. Used to discuss in a class by the author or the author's body and presentations at meetings or conferences and participant approval;

  2. Used for internal training by the author's company;

  3. Distribution to colleagues for the use of their research;

  4. Used in preparation for further author's works;

  5. Included in a thesis or dissertation;

  6. Partial or extra reuse of articles in other works (with full acknowledgment of the last item);

  7. Prepare derivatives (other than for commercial purposes);

  8. Post voluntarily on a website opened by the author or approve the author for scientific purposes (follow CC with a SA License).