Dự báo giá trị chịu rủi ro (VaR): Nghiên cứu từ các quốc gia Châu Á mới nổi=Forecasting Value at Risk: Evidence from Emerging Economies in Asia

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Trung Nghĩa Hoàng, Trung Thành Lê, Thị Ngân Nguyễn

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Tạp chí Phát triển Khoa học & Công nghệ: Khoa học - Kinh tế - Luật và Khoa học Quản lý, 2018

Mô tả vật lý: tr.77

Bộ sưu tập: Metadata

ID: 581476

Thêm vào giỏ

In this paper, various Value-at-Risktechniques are applied to stock indices of 9 Asianemerging financial markets. The results from ourselected models are then backtested by UnconditionalCoverage, Independence, Joint Tests ofUnconditional Coverage and Independence and Baseltests to ensure the quality of Value-at-Risk (VaR)estimates. The main conclusions are: (1) Timevaryingvolatility is the most important characteristicof stock returns when modelling VaR
(2) Financialdata is not normally distributed, indicating that thenormality assumption of VaR is not relevant
(3)Among VAR forecasting approaches, the backtestingbased on in- and out-of-sample evaluations confirmsits superiority in the class of GARCH models
Historical Simulation (HS), Filtered HistoricalSimulation (FHS), RiskMetrics and Monte Carlowere rejected because of its underestimation (for HSand RiskMetrics) or overestimation (for the FHS andMonte Carlo)
(4) Models under student’s t and skewstudent’s t distribution are better in taking intoaccount financial data’s characters
and (5)Forecasting VaR for futures index is harder than forstock index. Moreover, results show that there is noevidence to recommend the use of GARCH (1,1) toestimate VaR for all markets. In practice, the HS andRiskMetrics are popularly used by banks for largeportfolios, despite of its serious underestimations ofactual losses. These findings would be helpful forfinancial managers, investors and regulators dealingwith stock markets in Asian emerging economies.Trong bài nghiên cứu này, chúng tôi ápdụng nhiều kỹ thuật tính giá trị chịu rủi ro (VaR)của 9 chỉ số chứng khoán của các quốc gia Châu Ámới nổi. Kết quả từ các mô hình sau đó được kiểmtra lùi bằng các phương pháp như UnconditionalCoverage, Independence, Joint Tests ofUnconditional Coverage và Independence, Basel đểđảm bảo chất lượng của các ước tính VaR. Các kếtquả chính của nghiên cứu là: (1) Biến động thay đổitheo thời gian là đặc điểm quan trọng nhất của tỷsuất sinh lời chứng khoán khi mô hình hóa VaR
(2)Các số liệu tài chính không có phân phối chuẩn, hàmý rằng giả định phân phối chuẩn của VaR là khôngphù hợp
(2) Trong số các phương pháp dự báoVaR, kết quả kiểm tra lùi trong và ngoài mẫu chothấy các mô hình GARCH có độ chính xác vượt trội
Phương pháp Historical Simulation (HS), FilteredHistorical Simulation (FHS), RiskMetrics và MonteCarlo bị bác bỏ do dự báo quá cao (HS varRiskMetrics) hoặc dự báo quá thấp (FHS và MonteCarlo)
(4) Các mô hình có phân phối student’s t vàstudent’s t lệch tích hợp các đặc điểm của số liệu tàichính tốt hơn
và (5) Dự báo VaR đối với các chỉ sốtương lai khó hơn dự báo chỉ số chứng khoán. Ngoàira, kết quả cũng cho thấy không có cơ sở để khuyếnnghị dùng GARCH(1,1) để ước tính VaR cho tất cảcác thị trường. Trên thục tế, HS và RiskMetricsđược các ngân hàng sử dụng phổ biến đối với cácdanh mục lớn mặc dù các phương pháp này dự báotổn thất thực sự quá thấp. Những kết luận này sẽgiúp các nhà quản lý, đầu tư tài chính và cơ quanlập pháp quản lý tốt hơn thị trường chứng khoáncủa các quốc gia Châu Á mới nổi.

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR