So sánh mô hình dự báo Autoarima và học sâu LSTM cho dự báo giá cổ phiếu trên thị trường chứng khoán Việt Nam bằng Python=A comparison of Autoarima and deep learning LSTM models for stock price forecasting in the Vietnamese stock market using Python

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Đức Tài Đỗ, Thị Thu Hồng Ngô, Viết Giang Trịnh

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Nghiên cứu Tài chính Kế toán, 2024

Mô tả vật lý: tr.48-53

Bộ sưu tập: Metadata

ID: 241832

Thêm vào giỏ

In business production activities and government policy making, forecasting time series data is very necessary. Through an overview of previous research works, statistical models such as GDP, consumer price index, inflation, and stock prices have used time series data for forecasting. However, stock prices are very complex nonlinear data, affected by many factors, so forecasting is very difficult. Moreover, previous studies, especially in Vietnam, have rarely found the use of machine learning and deep learning designed by Python with available support packages to forecast economic variables. This article focuses on the comparative evaluation of ARIMA and LSTM deep learning models to forecast the VN-index index on the Ho Chi Minh Stock Exchange (HSX) from January 11, 2010 to August 11, 2024. The evaluation results of the deep learning model show that the deviation rate from reality is the lowest. From there, it is recommended that businesses, investors, and policymakers increase the use of technology in forecasting to support decision-making.Trong hoạt động sản xuất kinh doanh và hoạch định chính sách của chính phủ thì dự báo dữ liệu chuỗi thời gian là rất cần thiết. Qua tổng quan các công trình nghiên cứu trước, các mô hình thống kê như GDP, chỉ số giá tiêu dùng, lạm phát, giá cổ phiếu đã sử dụng dữ liệu chuỗi thời gian để dự báo. Tuy nhiên, giá cổ phiếu là dữ liệu phi tuyến rất phức tạp, bị ảnh hưởng bởi nhiều yếu tố cho nên dự báo là rất khó khăn. Hơn nữa, các nghiên cứu trước đặc biệt ở Việt Nam rất ít được tìm thấy việc sử dụng máy học và học sâu được thiết kế bởi Python với các gói hỗ trợ có sẵn để dự báo biến số kinh tế. Bài viết này tập trung vào đánh giá so sánh các mô hình ARIMA và học sâu LSTM dự báo cho chỉ số VN-index trên sàn giao dịch chứng khoán Hồ Chí Minh (HSX) từ 11/1/2010 đên 11/08/2024. Kết quả đánh giá mô hình học sâu cho thấy tỷ lệ sai lệch so với thực tế là thấp nhất. Từ đó, khuyến nghị các doanh nghiệp, các nhà đầu tư và nhà hoạ

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR