Claim Missing Document
Check
Articles

Found 26 Documents
Search

Penerapan Model ARIMA-ARCH untuk Meramalkan Harga Saham PT. Indofood Sukses Makmur Tbk Yulvia Fitri Rahmawati; Etik Zukhronah; Hasih Pratiwi
Jurnal Inovasi Bisnis dan Kewirausahaan Vol 3 No 3 (2021): Business Innovation and Entrepreneurship Journal (August)
Publisher : Entrepreneurship Faculty, Universitas Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (380.444 KB) | DOI: 10.35899/biej.v3i3.307

Abstract

Abstract– The stock price is the value of the stock in the market that fluctuates from time to time. Time series data in the financial sector generally have quite high volatility which can cause heteroscedasticity problems. This study aims to model and to predict the stock price of PT Indofood Sukses Makmur Tbk using the ARIMA-ARCH model. The data used is daily stock prices from 2nd June 2020 to 15th February 2021 as training data, while from 16th February 2021 to 1st March 2021 as testing data. ARIMA-ARCH model is a model that combines Autoregressive Integrated Moving Average (ARIMA) and Autoregressive Conditional Heteroscedasticity (ARCH), which can be used to overcome the residues of the ARIMA model which are indicated to have heteroscedasticity problems. The result showed that the model that could be used was ARIMA(1,1,2)-ARCH(1). This model can provide good forecasting result with a relatively small MAPE value of 0.515785%. Abstrak– Harga saham adalah nilai saham di pasar yang berfluktuasi dari waktu ke waktu. Data runtun waktu di sektor keuangan umumnya memiliki volatilitas cukup tinggi yang dapat menyebabkan masalah heteroskedastisitas. Penelitian ini bertujuan untuk memodelkan dan meramalkan harga saham PT Indofood Sukses Makmur Tbk menggunakan model ARIMA-ARCH. Data yang digunakan adalah harga saham harian dari 2 Juni 2020 hingga 15 Februari 2021 sebagai data training, sedangkan dari 16 Februari 2021 hingga 1 Maret 2021 sebagai data testing. Model ARIMA-ARCH merupakan suatu model yang menggabungkan Autoregressive Integrated Moving Average (ARIMA) dan Autoregressive Conditional Heteroscedasticity (ARCH), yang dapat digunakan untuk mengatasi residu dari model ARIMA yang terindikasi memiliki masalah heteroskedastisitas. Hasil penelitian menunjukkan bahwa model yang dapat digunakan adalah ARIMA(1,1,2)-ARCH(1). Model tersebut mampu memberikan hasil peramalan yang baik dengan perolehan nilai MAPE yang relatif kecil yaitu 0,515785%.
Model ARIMA-GARCH Pada Peramalan Harga Saham PT. Jasa Marga (Persero) Fransisca Trisnani Ardikha Putri; Etik Zukhronah; Hasih Pratiwi
Jurnal Inovasi Bisnis dan Kewirausahaan Vol 3 No 3 (2021): Business Innovation and Entrepreneurship Journal (August)
Publisher : Entrepreneurship Faculty, Universitas Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (420.185 KB) | DOI: 10.35899/biej.v3i3.308

Abstract

Abstract– PT Jasa Marga is a great reputation company, the leader in comparable businesses, has a steady income, and paying dividends consistently. This paper aims to find the best model to forecast stock price of PT Jasa Marga using ARIMA-GARCH. The data used is daily stock price of PT Jasa Marga from March 2020 to March 2021. Autoregressive Integrated Moving Average (ARIMA) is a method that can be used to forecast stock prices. However, an economical data tend to have heteroscedasticity problems, one of the methods used to overcome them is Generalized Autoregressive Conditional Heteroskedasticity (GARCH). Future stock price of PT Jasa Marga is forecasted with ARIMA-GARCH model. The data is modeled with ARIMA first, if there is heteroscedasticity, combine the model with GARCH model. The result of this study indicated that ARIMA (1, 1, 1) – GARCH (2, 2) is the best model, with MAPE 1,5647 Abstrak– PT Jasa Marga adalah perusahaan yang reputasinya baik, terdepan di perusahaan-perusahaan sejenis, stabil pendapatannya, dan pembayaran devidennya konsisten. Paper ini bertujuan untuk mencari model terbaik dalam meramalkan harga saham PT Jasa Marga menggunakan ARIMA-GARCH. Data harga saham yang diolah yaitu data sekunder dari PT Jasa Marga pada Maret 2020 hingga Maret 2021. Autoregressive Integrated Moving Average (ARIMA) sebagai metode yang dapat dimanfaatkan guna meramalkan harga saham. Akan tetapi, data tentang ekonomi cenderung memiliki masalah heteroskedastisitas, metode yang umum dipakai untuk mengatasinya adalah Generalized Autoregressive Conditional Heteroskedasticity (GARCH). Harga saham PT Jasa Marga diramalkan dengan model ARIMA-GARCH. Data terlebih dahulu dimodelkan dengan ARIMA, jika didapati adanya heteroskedastisitas, maka model tersebut dikombinasikan dengan GARCH. Penelitian ini menghasilkan ARIMA (1,1,1)-GARCH(2,2) sebagai model terbaik dengan MAPE 1,5647.
Analisis Data Panel pada Tingkat Pengangguran Terbuka Kabupaten/Kota di Pulau Jawa Hasih Pratiwi; Ardina Nilam Prawastyorini; Sugiyanto Sugiyanto
Jurnal Matematika, Statistika dan Komputasi Vol. 16 No. 1 (2019): JMSK, July, 2019
Publisher : Department of Mathematics, Hasanuddin University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (362.724 KB) | DOI: 10.20956/jmsk.v16i1.6713

Abstract

Unemployment rate is the percentage of the number of unemployed to the total labor force, it has become some problems in economic development. The aim of this study is to choose the best model between common, fixed, and random effects in modeling open unemployment rate of regency/city in Java. It based on open unemployment rate with several influence factors in Java Island 2010-2016 which are panel data types. The best model choosen based on the results of the Chow test and Hausman test. The fixed effect model was obtained as the best model with a value of  79,26 percent.
Sentiment Analysis Using Maximum Entropy on Application Reviews (Study Case: Shopee on Google Play) Ulinnuha Rhohmawati; Isnandar Slamet; Hasih Pratiwi
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol 5, No 1 (2019): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (287.386 KB) | DOI: 10.26555/jiteki.v5i1.13087

Abstract

Shopee was one of the e-commerce application that could found on Google Play. The amount of Shopee application reviews on Google Play continues to grow over time. These make the company trying to get the overall information from all reviews because it would take a long time to read each of the reviews on Google Play. Therefore analysis was used using text mining. One part of text mining was sentiment analysis that applied the maximum entropy method to classification. Based on the results of the analysis found an accuracy of 97.32%. By using the maximum entropy method it could be concluded that word association obtained related to “application”, “promo”, “satisfy”, and “discount” for positive sentiment. Meanwhile for negative sentiment, the reviewers of Shopee application on Google Play were related to “problematic”, “login”, “old”, “verification”, and “expensive”. The results of this research in Indonesian.
Peramalan curah hujan di kota bandung menggunakan singular spectrum analysis Tri Kartika Febrianti; Winita Sulandari; Hasih Pratiwi
Jurnal Ilmiah Matematika Vol 8, No 2 (2021)
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/konvergensi.v0i0.21461

Abstract

Curah hujan merupakan fenomena alam yang selalu terjadi di Indonesia setiap tahunnya. Fenomena ini bisa saja menyebabkan bencana seperti banjir dan tanah longsor. Adanya peramalan sangat dibutuhkan sebagai bentuk peringatan dini mengenai kondisi di waktu yang akan datang. Singular Spectrum Analysis (SSA) merupakan suatu teknik analisis deret waktu dan peramalan. SSA bertujuan untuk menguraikan deret waktu asli menjadi sejumlah kecil komponen yang dapat diinterpretasikan menjadi tren, osilasi dan noise. Tujuan dari penelitian ini yaitu menyajikan model peramalan curah hujan di Kota Bandung menggunakan metode Singular Spectrum Analysis (SSA). Berdasarkan penelitian ini, diketahui bahwa data curah hujan di Kota Bandung memiliki pola musiman. Penentuan window length (L) dilakukan dengan trial and error, yang dalam kasus ini diperoleh window length 17. Melalui dekomposisi dan rekonstruksi dengan window length 17 diperoleh 4 pengelompokan, yaitu satu kelompok tren dan tiga kelompok musiman. Pada penelitian ini digunakan RMSE untuk mengukur kesalahan hasil peramalan. Berdasarkan hasil pengujian dengan metode Singular Spectrum Analysis (SSA) diperoleh RMSE sebesar 167,510.
The Analysis of Flipped Learning Model Based on Geogebra and PBL on Mathematics Nur Rohman; Budiyono Budiyono; Hasih Pratiwi
Budapest International Research and Critics in Linguistics and Education (BirLE) Journal Vol 5, No 1 (2022): Budapest International Research and Critics in Linguistics and Education, Februa
Publisher : BIRCU

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33258/birle.v5i1.3914

Abstract

This research is motivated by the low learning outcomes of students' mathematics with an average national exam of around 55 low, the low learning outcomes are influenced by one of various factors including learning models or methods which sometimes do not support mathematics. Based on this, an alternative learning model is needed to provide understanding both in terms of cognitive and psychomotor. In this case, the geogebra-based flipped learning model helps understand the theory and practice of mathematics. This study aims to determine the regression between flipped learning and problem based learning (PBL) learning models in learning outcomes using geogebra software. This research is in SMPN/SMPT/SMP with the subject of class VIII students. The type of research is an experiment with a 3x3 factorial design with a random sampling technique, each of which is taught using the Flipped learning and problem based learning (PBL) models. Collecting data with tests and questionnaires, while the data analysis technique using inferential analysis of MANOVA and two-way ANOVA through prerequisite, balance, hypothesis and further tests. The conclusion is the knowledge aspect, the GeoGebra flipped learning (FLG) learning model provides better knowledge aspects than the Flipped learning (FL) and problem-based learning (PBL) learning models. The FL learning model provides better knowledge aspects than problem based learning (PBL). Meanwhile, in terms of skills, Flipped learning (FL) learning model provides better skill aspects than problem based learning (PBL); in the aspect of high interpersonal communication knowledge have better knowledge aspects than students with moderate interpersonal communication and low interpersonal communication. Students with moderate interpersonal communication have better aspects of knowledge than students with low interpersonal communication. Meanwhile, in the aspect of skills, students with high interpersonal communication have better skill aspects than students with moderate interpersonal communication and low interpersonal communication. Students with moderate interpersonal communication have better skill aspects than students with low interpersonal communication.
Back Matter Vol 2 No 2 Hasih Pratiwi
Indonesian Journal of Applied Statistics Vol 2, No 2 (2019)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v2i2.38496

Abstract

Back Matter Vol 4 No 2 2021 Hasih Pratiwi
Indonesian Journal of Applied Statistics Vol 4, No 2 (2021)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v4i2.56840

Abstract

Peramalan Banyak Pengunjung Pantai Pandasimo Bantul Menggunakan Regresi Runtun Waktu dan Seasonal Autoregressive Integrated Moving Average Exogenous Tito Tatag Prakoso; Etik Zukhronah; Hasih Pratiwi
Indonesian Journal of Applied Statistics Vol 4, No 1 (2021)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v4i1.45795

Abstract

Forecasting is a ways to predict what will happen in the future based on the data in the past. Data on the number of visitors in Pandansimo beach are time series data. The pattern of the number of visitors in Pandansimo beach is influenced by holidays, so it looks like having a seasonal pattern. The majority of Indonesian citizens are Muslim who celebrate Eid Al-Fitr in every year. The determination of Eid Al-Fitr does not follow the Gregorian calendar, but based on the Lunar calendar. The variation of the calendar is about the determination of Eid Al-Fitr which usually changed in the Gregorian calendar, because in the Gregorian calendar, Eid Al-Fitr day will advance one month in every three years. Data that contain seasonal and calendar variations can be analyzed using time series regression and Seasonal Autoregressive Integrated Moving Average Exogenous  (SARIMAX) models. The aims of this study are to obtain a better model between time series regression and SARIMAX and to forecast the number of Pandansimo beach visitors using a better model. The result of this study indicates that the time series regression model is a better model. The forecasting from January to December 2018 in succession are 13255, 6674, 8643, 7639, 13255, 8713, 22635, 13255, 13255, 9590, 8549, 13255 visitors.Keywords: time series regression, seasonal, calendar variations, SARIMAX, forecasting
Front Matter Vol 1 No 1 Hasih Pratiwi
Indonesian Journal of Applied Statistics Vol 1, No 1 (2018)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v1i1.24684

Abstract