본문 바로가기
[학술저널]

  • 학술저널

김유일 신은경 홍태호

UCI(KEPA) : I410-ECN-0101-2009-326-014511482

표지

북마크 2

리뷰 0

초록

This study presents the comparative analysis of data mining performance for the prediction of stock price index using neural networks and support vector machine. The prediction of stock price index was performed on the basis of technical analysis using technical indicators which is able to find the change of the present and future prices in the market. Neural networks have a few problems such as the lack of explanation and over-fitting although their outstanding performance in the financial prediction area. On the other side, SVM is capable of generalizing the model because it can be explained mathematically and minimize the structured risk.
In this study, we predicted the stock price index using neural networks and SVM and compared their performance with the prediction performance of discriminant analysis and logit, called statistical techniques. For the comparison of performance among each models, KOSPI 200 and S&P 500 index are utilized to experiments and the results are tested statistically. In addition, we analyzed the experimental results considering the characteristics of data in Korea and US.

목차

Abstract

Ⅰ. 서론

Ⅱ. SVM (Support Vector Machine)

Ⅲ. 주가지수 예측에 관한 문헌 연구

Ⅳ. 연구모형

Ⅴ. 연구모형의 실증 분석

Ⅵ. 결론

참고문헌

키워드

저자키워드
등록된 정보가 없습니다.

참고문헌(0)

리뷰(0)

도움이 되었어요.0

도움이 안되었어요.0

첫 리뷰를 남겨주세요.
DBpia에서 서비스 중인 논문에 한하여 피인용 수가 반영됩니다.
인용된 논문이 DBpia에서 서비스 중이라면, 아래 [참고문헌 신청]을 통해서 등록해보세요.
Insert title here