Peramalan Volume Transaksi Saham Pada Most Menggunakan Metode Autoregressive Integrated Moving Average

Najmi, Rafi Naufal (2024) Peramalan Volume Transaksi Saham Pada Most Menggunakan Metode Autoregressive Integrated Moving Average. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5006211083-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Laporan kerja praktik ini disusun berdasarkan pengalaman dan kegiatan yang dilakukan di PT Mandiri Sekuritas Jakarta. Fokus utama dari kerja praktik adalah melakukan peramalan terhadap volume transaksi saham pada platform Mandiri Online Securities Trading (MOST) menggunakan metode Autoregressive Integrated Moving Average (ARIMA). Data yang digunakan merupakan volume transaksi harian saham periode 11 September 2023 hingga 12 Juni 2024. Analisis dimulai dengan pemeriksaan stasioneritas data, transformasi Box-Cox, dan identifikasi orde model melalui grafik ACF dan PACF. Hasil analisis menunjukkan bahwa model ARIMA terbaik adalah AR(2), yang mampu menghasilkan nilai Mean Absolute Percentage Error (MAPE) sebesar 15,48% setelah dilakukan imputasi data. Peramalan volume transaksi selama satu bulan ke depan juga dibandingkan dengan interval kepercayaan untuk mengevaluasi tingkat ketidakpastian prediksi. Hasil penelitian ini memberikan gambaran yang bermanfaat bagi perusahaan dalam memahami pola aktivitas pasar serta sebagai pertimbangan dalam pengambilan keputusan strategis di bidang sekuritas.
====================================================================================================================================
This internship report is based on activities and experiences during the practical work period at PT Mandiri Sekuritas Jakarta. The main focus of the internship was forecasting stock transaction volume on the Mandiri Online Securities Trading (MOST) platform using the Autoregressive Integrated Moving Average (ARIMA) method. The data used consisted of daily stock transaction volumes from September 11, 2023, to June 12, 2024. The analysis began with stationarity testing, Box-Cox transformation, and model order identification through ACF and PACF plots. The results indicated that the best-fitting model was AR(2), which yielded a Mean Absolute Percentage Error (MAPE) of 15.48% after data imputation. The forecasted transaction volumes for the following month were also evaluated using confidence intervals to assess prediction uncertainty. This study provides valuable insights for the company in understanding market activity patterns and supporting strategic decision-making in the securities sector.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	ARIMA, peramalan, MOST, volume transaksi saham, stasioneritas, MAPE, ARIMA, forecasting, MOST, stock transaction volume, stationarity, MAPE
Subjects:	H Social Sciences > HA Statistics > HA30.3 Time-series analysis
Divisions:	Faculty of Mathematics, Computation, and Data Science > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Rafi Naufal Najmi
Date Deposited:	04 Aug 2025 02:49
Last Modified:	04 Aug 2025 02:49
URI:	http://repository.its.ac.id/id/eprint/126371

Actions (login required)

View Item