Peramalan Terkonfirmasi COVID 19 Di Kawasan Surabaya Menggunakan Metode ARIMA

Batry, Titanio Meiga (2023) Peramalan Terkonfirmasi COVID 19 Di Kawasan Surabaya Menggunakan Metode ARIMA. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05211740000047-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (3MB) | Request a copy

Abstract

COVID-19 merupakan penyakit baru yang berbeda dengan penyakit lain yang disebabkan oleh virus corona, seperti Severe Acute Respiratory Syndrome (SARS) dan Middle East Respiratory Syndrome (MERS). Virus ini menyebar dengan cepat dan dapat meningkat secara eksponensial. Saat ini, belum ada terapi atau vaksin yang terbukti untuk mengobati atau mencegah COVID-19, tetapi pemerintah, WHO, dan mitra mereka terus bekerja untuk mengoordinasikan perkembangan pesat obat pencegahan. Virus corona mulai menyerang penduduk di Indonesia pada 2 Maret 2020 dan terus menular hingga saat ini. Protokol kesehatan dan peraturan jaga jarak terus berjalan, sehingga diusahakan virus corona tidak akan menginfeksi lebih banyak orang di masa depan. Kesehatan fasilitas medis juga harus menjadi perhatian utama di masa pandemi ini. Banyak pasien yang tidak tertangani karena meningkatnya jumlah kasus positif Covid-19. Pada penelitian ini akan dilakukan peralaman jumlah kasus pasien yang terkonfirmasi Covid-19 di salah satunya pada kota Surabaya. Tujuan pada tugas akhir ini membuat peramalan jumlah terkonfirmasi positif Covid-19 di Kota Surabaya serta mengetahui nilai persentase error hasil peramalan terkonfirmasi Covid-19 di Kota Surabaya. Metode peramalan yang akan digunakan pada tugas akhir ini adalah metode Autoregressive Integrated Moving Average (ARIMA).ARIMA adalah penggabungan dari metode moving average dan metode autoregressive yakni suatu metode peramalan data runtun ketika memanfaatkan data historis dan data sekarang untuk menghasilkan peramalan jangka pendek yang akurat. Metode ini juga dapat melacak fluktuasi data dengan lebih baik dibandingkan dengan metode lainnya. Manfaat yang dapat dihasilkan dengan adanya tugas akhir ini adalah sebagai bahan pertimbangan dalam pengambilan keputusan untuk melakukan perencanaan kebutuhan untuk menanggulangi kasus Covid-19 di Kawasan Surabaya pada periode yang akan datang. Metode yang digunakan pada riset ini adalah metode ARIMA dengan bahasa pemograman R dengan menggunakan aplikasi yaitu R Studio. Namun untuk pengujian stasioneritas data sebelum melakukan peramalan metode ARIMA menggunakan aplikasi lain seperti IBM SPSS Statistic 26 dan Eviews 12 SV. Dari hasil peramalan terkonfirmasi positif Covid-19 Surabaya, peramalan model ARIMA yang terbaik dengan data pelatihan dalam penilitian ini adalah model ARIMA (0,1,2). Peramalan jumlah terkonfirmasi positif Covid-19 Surabaya dengan tingkat kesalahan yang cukup wajar yaitu SMAPE sebesar 35,9016%. Model ARIMA (0,1,2) juga lolos dari kedua pengujian ARIMA. Ada bebarapa model ARIMA yang lolos uji selain model ARIMA (0,1,2) adalah model ARIMA (2,1,1) dan model ARIMA (2,1,3). Nilai SMAPE model ARIMA (2,1,1) sebesar 35,9021% dan model ARIMA (2,1,3) sebesar 37,6350%. Bila dibandingkan metode lain dengan Naive seperti Mean Model dan Moving Average(Single dan Weighted) dengan tiap model ARIMA yang lolos tahap uji, SMAPE peramalan model ARIMA (0,1,2) memiliki kinerja 1,7334% lebih baik dibanding dengan Mean Model dan 0,1683% lebih baik dibanding dengan Single Moving Average, namun kinerja di bawah 1,7623% dengan Weighted Moving Average. SMAPE peramalan model ARIMA (2,1,1) memiliki kinerja 1,7329% lebih baik dibanding dengan Mean Model dan 0,1678% lebih baik dibanding dengan Single Moving Average, namun kinerja di bawah 1,7628% dengan Weighted Moving Average. SMAPE peramalan model ARIMA (2,1,3) memiliki kinerja 0,5423% lebih baik dibanding dengan Mean Model, namun kinerja di bawah 1,0228% dengan Single Moving Average dan 2,9534% dengan Weighted Moving Average. Jadi SMAPE model ARIMA (0,1,2) dan model ARIMA (2,1,1) memiliki kinerja lebih baik dibanding dengan Mean Model dan Single Moving Average, namun kinerja di bawah dengan Weighted Moving Average
==================================================================================================================================
COVID-19 is a new disease that is different from other diseases caused by corona viruses, such as Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS). This virus spreads quickly and can increase exponentially. Currently, there are no proven therapeutics or vaccines to treat or prevent Covid-19, but governments, WHO, and their partners are continuing to work to coordinate rapid treatment prevention. The corona virus began to attack residents in Indonesia on March 2 2020 and continues to spread to this day. Health protocols and social distancing regulations are continuing, so that the corona virus being attempted will not infect more people in the future. Medical health facilities must also be a major concern during this pandemic. Many patients are not being treated because of the increasing number of positive cases of Covid-19. In this study, an estimate will be made of the number of confirmed cases of Covid-19 in one of them in the city of Surabaya. The aim of this final project is to forecast the number of positive confirmed Covid-19 in the city of Surabaya and to know the proportion of errors in the forecasting results of confirmed Covid-19 in the city of Surabaya. The forecasting method that will be used in this final project is Autoregressive Integrated Moving Average method (ARIMA). ARIMA is a combination of the moving average method and the autoregressive method, which is a method for forecasting sequential data when utilizing historical data and current data to produce accurate short-term forecasts. This method can also track data fluctuations better than other methods. The benefits that can be generated from this final project are used as material for consideration in making decisions to plan needs to deal with Covid-19 cases in the Surabaya area in the coming period. The method used in this study is the ARIMA method with the R programming language using an application, namely R studio. However, to test the stationarity of the data before forecasting the ARIMA method, other applications such as IBM SPSS Statistics 26 and Eviews 12 SV are used. From the forecasting results confirmed positive for Covid-19 Surabaya, the best ARIMA model forecasting with training data in this research is the ARIMA model (0,1,2). Forecasting the number of positive confirmed Covid-19 Surabaya with a fairly reasonable error rate, namely SMAPE of 35.9016%. The ARIMA model (0,1,2) also passes both ARIMA tests. There are several ARIMA models that pass the test besides the ARIMA model (0,1,2), namely the ARIMA model (2,1,1) and the ARIMA model (2,1,3). The SMAPE value for the ARIMA model (2,1,1) is 35.9021% and for the ARIMA model (2,1,3) is 37.6350%. When compared to other methods with Naive such as the Mean Model and Moving Average (Single and Weighted) with each ARIMA model passing the test phase, the SMAPE forecasting ARIMA model (0,1,2) has a performance of 1.7334% better than the Mean Model and 0.1683% better than the Single Moving Average, but below the performance of 1.7623% with the Weighted Moving Average. SMAPE ARIMA forecasting model (2,1,1) has a performance of 1.7329% better than the Mean Model and 0.1678% better than the Single Moving Average, but the performance is below 1.7628% with the Weighted Moving Average. SMAPE ARIMA forecasting model (2,1,3) has a performance of 0.5423% better than the Mean Model, but the performance is below 1.0228% with the Single Moving Average and 2.9534% with the Weighted Moving Average. So the SMAPE ARIMA model (0,1,2) and ARIMA model (2,1,1) have better performance than the Mean Model and Single Moving Average, but the performance is below the Weighted Moving Average

Item Type:	Thesis (Other)
Uncontrolled Keywords:	COVID-19, Peramalan, ARIMA COVID-19, Forecasting, ARIMA
Subjects:	T Technology > T Technology (General) > T174 Technological forecasting
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Titanio Meiga Batry
Date Deposited:	03 Feb 2023 04:12
Last Modified:	03 Feb 2023 04:12
URI:	http://repository.its.ac.id/id/eprint/96137

Actions (login required)

View Item