Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods
碩士 === 國立高雄第一科技大學 === 金融營運所 === 90 === This paper utilized the Back-propagation network(BPN)that input variables included technical indicators, original stock price information and price enveloping variables with different network parameters to establish the optimal stock price forecasting models. F...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2002
|
Online Access: | http://ndltd.ncl.edu.tw/handle/40412751799044155316 |
id |
ndltd-TW-090NKIT5667029 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-090NKIT56670292015-10-13T10:21:17Z http://ndltd.ncl.edu.tw/handle/40412751799044155316 Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods 運用股價原始資訊建構股價預測模型-類神經網路之應用 Wan-Hsin Liu 劉宛鑫 碩士 國立高雄第一科技大學 金融營運所 90 This paper utilized the Back-propagation network(BPN)that input variables included technical indicators, original stock price information and price enveloping variables with different network parameters to establish the optimal stock price forecasting models. Finally, models were tested to beat the market. The samples were the twenty largest listed electronic companies in terms of market value. Daily data were used from January 5,1995 to March 15,2002. The findings are summarized as follow: 1.Because of choosing the technical indicators and the definition of technical indicators’ period inappropriate for the study, the technical indicators model has lower accurately than the original information model. 2.The original information model has better investment performance than the technical indicators model and buy & hold strategy. Although this paper determines the best forecasting model, it would not have the best investment performance. The reasons include of the assumes that this paper set and the non-economical factors Gu-Ann Yang 楊筑安 2002 學位論文 ; thesis 71 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立高雄第一科技大學 === 金融營運所 === 90 === This paper utilized the Back-propagation network(BPN)that input variables included technical indicators, original stock price information and price enveloping variables with different network parameters to establish the optimal stock price forecasting models. Finally, models were tested to beat the market.
The samples were the twenty largest listed electronic companies in terms of market value. Daily data were used from January 5,1995 to March 15,2002.
The findings are summarized as follow:
1.Because of choosing the technical indicators and the definition of technical indicators’ period inappropriate for the study, the technical indicators model has lower accurately than the original information model.
2.The original information model has better investment performance than the technical indicators model and buy & hold strategy. Although this paper determines the best forecasting model, it would not have the best investment performance. The reasons include of the assumes that this paper set and the non-economical factors
|
author2 |
Gu-Ann Yang |
author_facet |
Gu-Ann Yang Wan-Hsin Liu 劉宛鑫 |
author |
Wan-Hsin Liu 劉宛鑫 |
spellingShingle |
Wan-Hsin Liu 劉宛鑫 Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods |
author_sort |
Wan-Hsin Liu |
title |
Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods |
title_short |
Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods |
title_full |
Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods |
title_fullStr |
Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods |
title_full_unstemmed |
Utilizing Stock Index Original Information on Forecasting Individual Stock Price by Artificial Neural Networks Methods |
title_sort |
utilizing stock index original information on forecasting individual stock price by artificial neural networks methods |
publishDate |
2002 |
url |
http://ndltd.ncl.edu.tw/handle/40412751799044155316 |
work_keys_str_mv |
AT wanhsinliu utilizingstockindexoriginalinformationonforecastingindividualstockpricebyartificialneuralnetworksmethods AT liúwǎnxīn utilizingstockindexoriginalinformationonforecastingindividualstockpricebyartificialneuralnetworksmethods AT wanhsinliu yùnyònggǔjiàyuánshǐzīxùnjiàngòugǔjiàyùcèmóxínglèishénjīngwǎnglùzhīyīngyòng AT liúwǎnxīn yùnyònggǔjiàyuánshǐzīxùnjiàngòugǔjiàyùcèmóxínglèishénjīngwǎnglùzhīyīngyòng |
_version_ |
1716828044377718784 |