Article Per Year (5 Year)
p-Index From 2021 - 2026
4.449
P-Index
This Author published in this journals
All Journal JURNAL ELEKTRO Indonesia Symposium on Computing Indonesian Journal on Computing (Indo-JC) IJoICT (International Journal on Information and Communication Technology) JOIN (Jurnal Online Informatika) JURNAL MEDIA INFORMATIKA BUDIDARMA Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI) JIPI (Jurnal Ilmiah Penelitian dan Pembelajaran Informatika) TEKTRIKA - Jurnal Penelitian dan Pengembangan Telekomunikasi, Kendali, Komputer, Elektrik, dan Elektronika Building of Informatics, Technology and Science Journal of Computer System and Informatics (JoSYC) Charity : Jurnal Pengabdian Masyarakat Jurnal Indonesia Sosial Teknologi eProceedings of Engineering LOGIC: Jurnal Penelitian Informatika
Deni Saepudin
Universitas Telkom Bandung, Indonesia
Arts Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Industrial & Manufacturing Engineering Law, Crime, Criminology & Criminal Justice Social Sciences

Published : 111 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 111 Documents
Search

 7   8   9   10   11 
Gabungan Metode Support Vector Machine (SVM) dan Equal Weight Portofolio (EWP) Untuk Pengambilan Keputusan Jual Beli Saham Novelya Nababan; Deni Saepudin; Aniq Atiqi Rohmawati
eProceedings of Engineering Vol 6, No 2 (2019): Agustus 2019
Publisher : eProceedings of Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Salah satu metode yang dapat digunakan untuk memprediksi pergerakan harga saham yaitu Support Vector Machine (SVM) dengan menggunakan kernel linear. Pada penelitian ini menggunakan tiga pendekatan sebagai inputan model, pendekatan pertama untuk inputan data diperoleh dari perhitungan inputan SVM dengan dua puluh dua atribut menggunakan data trading (open, high, low, dan close prices) yang dimana algoritma SVM akan mendapatkan prediksi dan menjadi nilai bobot, nilai bobot yang didapatkan digunakan untuk menghitung return portofolio, dan EWP memberi bobot yang sama kepada semua perusahaan, sedangkan mean variance (MV) akan mendapatkan bobot yang akan di masukkan ke dalam prediksi SVM pada semua perusahaan. Pada penelitian ini menggunakan data historis setiap perusahaan dari 2005 sampai 2018. Data ini digunakan untuk mempelajari pola yang pada akhirnya dapat memprediksi pergerakan harga saham dari setiap perusahaan. Kinerja Algoritma SVM + EWP menunjukkan hasil yang optimal dibandingkan dengan EWP tanpa SVM masih belum menunjukkan hasil yang optimal. Nilai maksimal yang diperoleh return portofolio SVM + EWP adalah 14.71%, return portofolio EWP tanpa SVM adalah 0.27%, dan return portofolio SVM + MV adalah 0.12%, dengan nilai rata-rata return portofolio masing-masing algoritma adalah 10.30%, 0.94%, 1.76%. Kata kunci:SVM, Equal Weight Portofolio, Mean Variance
Pembentukan Portofolio Saham Melalui Proses Clustering Kurva Harga Saham Hasil Cubic-Spline Faturachman Nugraha Sasmita; Deni Saepudin; Annisa Aditsania
eProceedings of Engineering Vol 6, No 2 (2019): Agustus 2019
Publisher : eProceedings of Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Pelaku transaksi saham seringkali mengalami kesulitan dalam menentukan waktu yang tepat untuk membeli atau menjual saham. Hal ini disebabkan karena waktu pembelian saham dapat menentukan keuntungan investasi pada suatu saham. Maka dari itu, diperlukan portofolio saham yang dapat mendiversifikasi harga saham sehingga dapat membantu pembeli maupun penjual saham dalam bertransaksi di pasar modal. Penelitian ini membahas mengenai pembuatan portofolio saham melalui clustering kurva harga saham yang berasal dari metode cubic spline. Cubic spline untuk menginterpretasikan data yang sudah direduksi. Metode clustering pada penelitian kali ini dipakai untuk mengelompokkan koefisien cubic-spline dan menghasilkan 2,3, dan 4 clustering saham yang pengelompokkannya digunakan dengan metode K-means. Selanjutnya dilakukan pembentukan portofolio saham dengan memilih satu perwakilan dari setiap clustering berdasarkan rata-rata return setiap saham. Penelitian ini menghasilkan portofolio dengan nilai risiko terendah untuk pembagian cluster menjadi 4 cluster sebesar 0.0598 jika dibandingkan dengan pembagian cluster menjadi 2 cluster sebesar 0.1049 dan 3 cluster sebesar 0.2396. Kata kunci : portofolio, saham, cubic-spline, k-means, clustering
Prediksi Arah Kenaikan Indeks Sektoral yang Berada Di Bursa Efek Indonesia (BEI) dengan Menggunakan Bayesian Network Benedikto Krisnandy Wijaya; Deni Saepudin; Aniq Atiqi Rohmawati
eProceedings of Engineering Vol 7, No 1 (2020): April 2020
Publisher : eProceedings of Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Abstrak Peramalan data khususnya pergerakan indeks merupakan suatu metode yang digunakan untuk membantu pengambilan keputusan dalam berinvestasi di pasar keuangan. Investasi saham sendiri dilakukan untuk meningkatkan aset pada masa depan. Dalam investasi juga harus mempertimbangkan hasil yang didapatkan atau biasa disebut return. Untuk bisa mengetahui pergerakan dan hubungannya di masa depan, dibutuhkan sebuah model untuk membantu meramalkan pergerakan harga saham. Dalam tugas akhir ini akan membahas tentang, bagaimana memprediksi arah kenaikan indeks sektoral yang berada di Bursa Efek Indonesia (BEI) dengan menggunakan Bayesian Network dan Algoritma Naïve Bayes. Indeks sektoral yang digunakan adalah data historis mingguan dari tahun 2000 sampai 2018 yang setiap sektornya berjumlah sebanyak 984 minggu yaitu 02 Januari 2000 sampai 27 Desember 2015 yang berjumlah 828 minggu sebagai data training dan data testing antara 03 Januari 2016 sampai 30 Desember 2018 yang berjumlah sebanyak 156 minggu. Metode yang digunakan untuk mengetahui pergerakan setiap indeks adalah menggunakan Algoritma Naïve Bayes Diskrit dan Kontinu. Setiap indeks diasumsikan saling bebas dan hanya berkaitan degan nilai kurs dollar Amerika Serikat. Dari hubungan tersebut digunakanlah Bayesian Network untuk menggambarkan topologinya. Setelah itu, menentukan metode yang terbaik untuk perhitungan prediksi dengan melihat tingkat akurasi dari setiap metode dengan confusion matrix. Indeks yang terkait diantaranya adalah JKAGRI, JKCONS, JKFINA, JKINFA, JKMING, JKPROP, dan JKTRAD terhadap Kurs Dollar Amerika Serikat (USD/IDR). Nilai rata-rata akurasi pada Naïve Bayes Diskrit adalah sebesar 60.71% untuk data training dan 55.43% untuk data testing. Sedangkan nilai rata-rata akurasi pada Naïve Bayes Kontinu adalah sebesar 51.28% untuk data testing. Sektor nilai tukar USD/IDR lebih mempengaruhi JKINFA pada Data Training, sedangkan pada Data Testing lebih mempengaruhi JKAGRI dan JKCONS. Kata kunci : Indeks Sektoral, Bayesian Network, Naïve Bayes Diskrit, Naïve Bayes Kontinu, Data Historis Abstract Data forecasting, especially index movements, is a method used to assist decision making in investing in financial markets. Own stock investment is carried out to increase assets in the future. In investment must also consider the results obtained or commonly called return. To be able to know the movements and relationships in the future, we need a model to help predict stock price movements. In this final project will discuss about, how to predict the direction of the rise in sectoral indices on the Indonesia Stock Exchange (IDX) using the Bayesian Network and Naïve Bayes Algorithm. The sectoral index used is weekly historical data from 2000 to 2018, with each sector totaling 984 weeks, namely 2 January 2000 to 27 December 2015 totaling 828 weeks as training data and testing data between 3 January 2016 and 30 December 2018, totaling 156 Sunday. The method used to determine the movement of each index is using the Discrete and Continuous Naïve Bayes Algorithm. Each index is assumed to be independent and only relates to the value of the US dollar exchange rate. From this connection Bayesian Network is used to describe the topology. After that, determine the best method for calculating predictions by looking at the accuracy of each method with the confusion matrix. Related indexes include JKAGRI, JKCONS, JKFINA, JKINFA, JKMING, JKPROP, and JKTRAD against the US Dollar Exchange Rate (USD / IDR). The average accuracy in the Discrete Naïve Bayes is 60.71% for training data and 55.43% for testing data. While the average value of accuracy in Continuous Naïve Bayes is 51.28% for testing data. The USD / IDR exchange rate sector has more influence on JKINFA in Data Training, while in Data Testing it has more influence on JKAGRI and JKCONS. Keywords: Sectoral Indices, Bayesian Networks, Naïve Bayes Discrete, Naive Bayes Continuous, Historical Data
Artificial Neural Network Untuk Prediksi Pergerakan Harga Saham Sektor Keuangan Dengan Melibatkan Data Google Trends Febry Triyadi; Deni Saepudin; Aniq Atiqi Rohmawati
eProceedings of Engineering Vol 7, No 2 (2020): Agustus 2020
Publisher : eProceedings of Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Abstrak Dalam dunia bisnis, memprediksi harga saham adalah salah satu tantangan yang sulit bagi para investor, sehingga banyak penelitian yang dilakukan untuk memprediksi harga saham. Google trends adalah grafik statistik pencarian web yang menampilkan informasi yang paling trend dan paling banyak menjadi perhatian orang menurut google pada kurun waktu tertentu. Banyak penelitian yang memanfaatkan data Google trends untuk memprediksi data time series, dikarenakan data Google trends itu selalu update dan mudah diakses. Pada penelitian ini, pergerakan harga saham dalam sektor keuangan dengan melibatkan data Google trends menggunakan metode Artificial neural network (ANN). Penelitian ini juga menyelidiki pengaruh Google trends terhadap data harga saham menggunakan metode Cointegration test dan Granger causality analysis. Data Google trends menjadi salah satu cara untuk mempertimbangkan hasil prediksi harga saham. Penelitian ini juga membandingkan dua prediksi menjadi tipe I dan tipe II diantaranya: tipe I prediksi pergerakan harga saham tanpa data Google trends dan tipe II prediksi pergerakan harga saham dengan data Google trends. Hasil menunjukan bahwa prediksi dengan penambahan data Google trends memberikan dampak yang sedikit lebih signifikan terhadap hasil prediksi dibandingkan prediksi tanpa Google trends. Kata kunci : Google trends, harga saham, Artificial neural network Abstract In the business world, predicting share prices is one of the difficult challenges for investors, so a lot of researches were done to predict share prices. Google trends is a graph of web search statistics that shows information which have the most popular information and attention of people according to Google at a certain time. Many studies use Google trends data to predict time series data, because Google trends data is always up to date and easily accessible. In this study, the movement of stock prices in the financial sector by involving Google trends data using the Artificial neural network (ANN) method. This study also investigates the effect of Google trends on stock price data using the Cointegration test and the Granger causality analysis methods. Google trends data is a way to consider the results of stock price predictions. This study also compares two predictions with Type I and Type II, which include: Type I is prediction of stock price movements without Google trends data and Type II is prediction of stock price movements with Google trends data. The results show that predictions with the addition of Google trend data have a slightly more significant impact on the results of the predictions than the predictions without Google trends. Keywords: Google trends, stock price, Artificial neural network
Optimasi Portofolio Saham Lq45 Dengan Mempertimbangkan Prediksi Return Menggunakan Metode Support Vector Regression (svr) Vina Putri Damartya; Deni Saepudin; Putu Harry Gunawan
eProceedings of Engineering Vol 8, No 5 (2021): Oktober 2021
Publisher : eProceedings of Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Optimasi portofolio saham dibutuhkan investor untuk mendapatkan hasil yang diharapkan. Untuk mendapatkan portofolio yang diharapkan, dibutuhkan prediksi guna untuk menghasilkan bobot yang optimal. Optimasi portofolio sudah dikembangkan sejak lama, namun biasanya hanya mempertimbangkan risiko dan nilai harapan. Berbeda dengan pendekatan sebelumnya, mengintegrasikan prediksi return pada model time series tradisional dalam pembentukan portofolio dapat meningkatkan kinerja model pengoptimalan portofolio asli. Machine learning telah menunjukkan keunggulan yang luar biasa dalam prediksi pasar saham, banyak peneliti menerapkan model-model ini dalam proses pembentukan portofolio dan menghasilkan hasil yang memuaskan, penelitian ini menggabungkan prediksi return dalam pembentukan portofolio dengan metode Support Vector Regression (SVR). Adapun data saham yang digunakan dalam tugas akhir ini adalah saham LQ45. Berdasarkan hasil pengujian, hasil dari prediksi return menggunakan metode Support Vector Regession (SVR) dievalusi menggunakan Root Mean Square Error (RMSE) mendapatkan nilai 0.34973. Portofolio yang mempertimbangkan prediksi return menghasilkan kinerja yang lebih baik dibandingkan Indeks LQ45 yang diukur berdasarkan nilai rata-rata return portofolio, standar deviasi dan sharpe ratio. Kata kunci : Optimasi portofolio, return, support vector regression, LQ45
Portfolio Optimization Based on Return Prediction and Semi Absolute Deviation (SAD) Gharyni Nurkhair Mulyono; Deni Saepudin; Aniq Atiqi Rohmawati
International Journal on Information and Communication Technology (IJoICT) Vol. 9 No. 1 (2023): June 2023
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v9i1.698

Abstract

A portfolio is a collection of investment financial assets managed by financial institutions or individuals. In investment activities, investors expect minimal loss risk and optimal stock portfolio weight to get maximum profit. Investors can monitor changes in stock index values to compare portfolio performance. This research has discussed how to build a portfolio based on stock datasets with the LQ45 index using return predictions from the artificial neural network (ANN) method with semi-absolute deviation (SAD). Furthermore, the portfolio is optimized by looking for weights that match it. After that, a comparison of portfolio performance was carried out using the Sharpe ratio (SR) method between the semi-absolute deviation (SAD) portfolio and the portfolio resulting from the formation of the equal weight (EW) portfolio. Portfolio performance with ANN prediction and SAD is better than equal-weight portfolios in terms of mean return, standard deviation, and sharpe ratio for portfolios with few stocks, namely 2 and 3 stocks. In addition, a portfolio with a higher number of stocks can make the portfolio value from the ANN close prediction algorithm process and the selection of weights based on SAD is better than portfolios with equal weight for each list of stocks in the portfolio.
Optimizing LQ45 Stock Portfolio To Maximize Sharpe Ratio Value Using LSTM Tasya Salsabila; Deni Saepudin; Aniq Atiqi Rohmawati
Journal of Computer System and Informatics (JoSYC) Vol 4 No 2 (2023): Februari 2023
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v4i2.3065

Abstract

Investment is an investment activity within a certain period with the hope of getting a profit. Things that need to be considered by investors when investing are not just yields (return), but investors need to consider the purpose of the investment and the investment period. This study optimizes the formation of portfolios by utilizing the predicted value of stock prices using LSTM. The test used five daily stock indices from LQ45, namely BBCA, BBRI, TLKM, UNVR, and BMIR, from April 2010 – April 2020. The portfolio was built using the Genetic Algorithm and Equal-Weight (EW) method. Portfolio of Genetic Algorithm and Equal-Weight (EW) without predictions used as a benchmark. The experimental results show that using the LSTM prediction and Genetic Algorithm can produce an optimal portfolio with the highest Sharpe ratio value at 1.3950.
Stock Industry Sector Prediction Based on Financial Reports Using Random Forest Zhafran, Kamil Elian; Saepudin, Deni
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5743

Abstract

This study aims to predict the stock industry sector on the Indonesia Stock Exchange (IDX) based on financial reports using the Random Forest method. Implementing this machine learning approach is crucial due to the complexity of financial data, which demands robust and adaptive methods for accurate predictions. The dataset comprises financial data from companies across 10 industrial sectors on the IDX, spanning 2010-2022, and includes 17 features from each financial report. Notably, there is an imbalance in the number of companies per sector, with sector B representing 14.76% and sector G only 1.98%. This imbalance introduces bias in data analysis, thus necessitating the application of the SMOTE oversampling method to address it. The research process involves data cleaning, splitting the data into 80% training and 20% testing sets, applying the SMOTE oversampling technique, and comparing predictions from imbalanced and balanced datasets. The Random Forest method is chosen for its capability to handle complex datasets for industrial sector classification. Evaluation results indicate that without oversampling, the model achieves an accuracy of 73.57%, precision of 74.29%, recall of 73.57%, and an F1-score of 73.51%. With oversampling, these metrics improve to an accuracy of 80.21%, precision of 81.34%, recall of 80.21%, and an F1-score of 80.45%.
Clustering-Based Stock Return Prediction using K-Medoids and Long Short-Term Memory (LSTM) Sofyan, Denny; Saepudin, Deni
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.5744

Abstract

This research focuses on predicting stock returns using the K-Medoids clustering method and the Long Short-Term Memory (LSTM) model. The primary challenge lies in forecasting stock prices, which are then converted into return predictions. Clustering is performed to group stocks with similar price movements, facilitating the preparation of data for training the LSTM model within each cluster. This issue is crucial for aiding investors in making more informed investment decisions by leveraging predictions within specific stock clusters. Through clustering with K-Medoids, based on average returns and return standard deviation, the LSTM model is trained to predict daily returns for each stock within different clusters using the average stock price in each cluster. The data is divided into training (2013-2019) and testing (2020-2022) datasets, with model evaluation conducted using Root Mean Square Error (RMSE). The implementation results indicate prediction performance measured by RMSE for each cluster, with Cluster 3 showing the best performance with a testing RMSE of 0.0300, while Cluster 4 exhibited the worst performance with an RMSE of 0.3995. In the formation of an equal weight portfolio, tested from May 2020 to January 2023, the portfolio value grew from 1 to 2.50, with an average return of 0.0014 and a return standard deviation of 0.0158, indicating potential gains with lower risk compared to the LQ45 index.
Application of Deep Reinforcement Learning for Stock Trading on The Indonesia Stock Exchange Saepudin, Deni; Rauf, Khalifatur
Jurnal Nasional Pendidikan Teknik Informatika: JANAPATI Vol. 14 No. 1 (2025)
Publisher : Prodi Pendidikan Teknik Informatika Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/janapati.v14i1.83775

Abstract

In the last couple of years, stock trading has gained so much popularity because of its promising returns. However, most investors do not pay attention to the risks of trading without analysis, which can lead to a big loss. Some to reduce these risks, try their luck with automated and pre-programmed trading systems, which are called Expert Advisors. The current study examines the application of DRL for automated assistance in trading with an emphasis on decision-making enhancement, particularly the use of DRL in order to realize high asset returns with a low risk of exposure. Concretely, the two applied DRL methods within this work are A2C and PPO. By systematic testing, the A2C method produced a Sharpe Ratio of 1.6009 with a cumulative return of 1.4468, while the PPO method achieved a Sharpe Ratio of 1.7628 with a cumulative return of 1.4767. These were fine-tuned for the most optimal learning rates, cut loss, and take profit ratios, thus showing great promise with the capability to tune up trading strategies and improve trading performances. The research leverages these DRL techniques, hence arriving at better trading strategies that balance profit and risk, while underlining the promise of advanced algorithms in automated stock trading.
Co-Authors Abdurrahman Muttaqiin Achmad Fadholy Achmad Rizal Aditya Firman Ihsan Adiwijaya Aisyah Aisyah Alberila Fraida Loceseima Putri Almaya Sofariah Andhika Rama Putra Anggia Parsaoran Exaudi Aniq Antiqi Rohmawati Aniq Atiqi Rohmawati Aniq Rohmawati Anjar Pratiwi Annas Wahyu Ramadhan Annisa Aditsania Annisa Resnianty Anton Sri Haryanto Arfananda, Muhammad Ghifari Arifin Dwi Kandar Saputro Ayunda Firsty Trisnowati Azizah , Nakhwa Benedikto Krisnandy Wijaya Caramoy, Senza Danar Satrio Aji Dara Ayu Lestari Defy Ayu Dewa Made Rai Widyadarma Diah Fitri Wulandari Diani Sarah Kamilial Diani Sarah Kamilial Didit Adytia Dimas Rizqi Guintana Dini Apriliani Lestari Dio Navialdy Egi Shidqi Rabbani Elvina Oktavia Erlina Febriani Esther Laura Christy Fadhlika Hadi Fahmi Muhamad Fauzi Farah Diba Faturachman Nugraha Sasmita Fazlur Rahman Amri Febry Triyadi Fhira Nhita Fikri Nur Hadiansyah Fitriaini Amalia Freyssenita Kanditami P Furqon Hidayat Gharyni Nurkhair Mulyono Ghufron, Sayid Giali Ghazali Gilang Rachman Perdana Gilang Rachman Perdana Hadyatma Dahna Marta Hario Adi Ghufron Herlansyah, Ridhwan Rifky Himatul Zulfa Husain Athfal Hidayat I Kamil Elian Zhafran Ihsan Hasanudin Irfan Fauzan Prasetyo Irma Palupi Isman Kurniawan Izzata Izzata Jondri Jondri Kaisa Sekaring Pertiwi Kautsar Abdillah Kemas Muslim Lhaksmana Khoirunnisa Ulayya Kuntjoro Adji Sidarto Lani Rohaeni Laode Muhammad Ali Al-Qomar Lesmana, Rangga Made Larita Ditakristy Mailia Putri Utamil Maulid Fathurachman, Rizaldi Mayriskha Isna Indriyani Mega Silvia Desvi Muhamad Aziz, Reihan Muhammad Fadhil Maulana Muhammad Iqbal Cholil Muhammad Rifqi Arrahim Natadikarta Muhammad Taufiq Raihan Nanda Putri Mintari Narestha Adi Pratama, Putu Agus Naufal Abdurrahman Burhani Nisrina Nur Faizah Novelya Nababan Novi Syafira, Muthia Nur Roza Fitriyana Putri Nuvaisiyah Putu Harry Gunawan Rahmi Putri Amalia Raisa Betha Meiliza Ratih Puspita Furi Rauf, Khalifatur Razaq, Kukuh Sanddi Reima Agustina Kusumawardani Reiza Krisnaviardi Resi Annisa Nur Reza Pratama Rian Febrian Umbara Ridhwan Rifky Herlansyah Rizaldi Maulid Fathurachman Rizq Athariq, Muhammad Sabilla Fitriyantini Saputra, Muhammad Ridho Semeidi Husrin Shabrina Nanggala Sheila Nur Fadhila Sofyan, Denny Sri Rezeki Hardiyanti Susy Sundari Syaifrijal Zirkon Radion Tasya Salsabila Tifani Intan Solihati Triandini Nurislamiaty Triyana Kadarisman Uggi Stivani Savitri Vina Putri Damartya Widyasari, Felicia Dina Yanuar Ishaq Zhafran, Kamil Elian