Peramalan Volume Transaksi Saham Nasabah Retail Perusahaan Mandiri Sekuritas Menggunakan Metode Autoregressive Integrated Moving Average

Siahaan, Keisha Alessandra Lynn (2024) Peramalan Volume Transaksi Saham Nasabah Retail Perusahaan Mandiri Sekuritas Menggunakan Metode Autoregressive Integrated Moving Average. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5006211087-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Penelitian ini bertujuan untuk meramalkan volume transaksi saham nasabah retail di PT Mandiri Sekuritas menggunakan metode Autoregressive Integrated Moving Average (ARIMA). Data yang digunakan adalah volume transaksi harian periode 1 Juli 2023 hingga 31 Juli 2024. Proses analisis meliputi pemeriksaan stasioneritas, identifikasi model, estimasi parameter, uji diagnostik, dan pemilihan model terbaik berdasarkan RMSE dan MAPE. Hasil penelitian menunjukkan bahwa data memerlukan transformasi Box-Cox untuk mencapai stasioneritas varian, namun sudah stasioner pada rata-rata. Model terbaik yang terpilih adalah ARIMA (1,0,1) dengan nilai RMSE 0,24 pada data training dan 0,22 pada data testing, serta MAPE 18,09% dan 15,7%. Model ini mampu memprediksi volume transaksi retail dengan tingkat kesalahan yang relatif rendah. Hasil peramalan diharapkan dapat menjadi acuan bagi manajemen risiko Mandiri Sekuritas dalam merencanakan strategi bisnis yang lebih efektif.
=====================================================================================================================================
This study aims to forecast the stock trading volume of retail customers at PT Mandiri Sekuritas using the Autoregressive Integrated Moving Average (ARIMA) method. The data analyzed consist of daily transaction volumes from July 1, 2023, to July 31, 2024. The analysis process includes stationarity testing, model identification, parameter estimation, diagnostic checking, and selecting the best model based on RMSE and MAPE metrics. The results indicate that the data required a Box-Cox transformation to achieve variance stationarity but were already stationary in terms of the mean. The best-performing model selected was ARIMA (1,0,1), with RMSE values of 0.24 for training data and 0.22 for testing data, and MAPE values of 18.09% and 15.7%, respectively. This model is capable of predicting retail transaction volumes with relatively low error rates. The forecast results are expected to serve as a reference for Mandiri Sekuritas’ risk management in planning more effective business strategies.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	stock trading volume, retail customers, ARIMA, forecasting, Mandiri Sekuritas, volume transaksi saham, nasabah retail, ARIMA, peramalan, Mandiri Sekuritas
Subjects:	Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Keisha Alessandra Lynn Siahaan
Date Deposited:	28 Jul 2025 08:12
Last Modified:	28 Jul 2025 08:12
URI:	http://repository.its.ac.id/id/eprint/121459

Actions (login required)

View Item