Applying Fuzzy time series for Stock Forecasting
碩士 === 崑山科技大學 === 企業管理研究所 === 93 === It is necessary to split historical data into equal interval while use traditional fuzzy time series, because the splitting is too subjective, and neglect the attribute of data distributed, it makes result appeared to uncertainty, this research hope propose a tec...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2005
|
Online Access: | http://ndltd.ncl.edu.tw/handle/3589p2 |
id |
ndltd-TW-093KSUT5121045 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-093KSUT51210452019-05-15T20:33:45Z http://ndltd.ncl.edu.tw/handle/3589p2 Applying Fuzzy time series for Stock Forecasting 運用模糊時間序列於股價預測 Jing-Ru Tzeng 曾靖儒 碩士 崑山科技大學 企業管理研究所 93 It is necessary to split historical data into equal interval while use traditional fuzzy time series, because the splitting is too subjective, and neglect the attribute of data distributed, it makes result appeared to uncertainty, this research hope propose a technology that have more effective and accurate, and could use it to be the fuzzy tool among numerous fuzzy time series methods. According to the characteristics of data distributed, New method that variation and non- regular of interval to be proposed of this research, and with number of people entrance register university of Alabama as an example, the result is superior to other methods while compare with others that have used 4 different types data distributed, normal, uniform, poisson, exponential. This result has verified the prediction error of the method that this research institute proposes is far lower than the traditional fuzzy time series, the prediction ability to verify this method is the best at present. This research so as to predict the technological method is applied to the stock price of Taiwan and predicted, it predicts the effect is better than the prediction of other methods too, the result of this real example proves the superiority of this research approach even more. Hai-Wen Lu 陸海文 2005 學位論文 ; thesis 65 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 崑山科技大學 === 企業管理研究所 === 93 === It is necessary to split historical data into equal interval while use traditional fuzzy time series, because the splitting is too subjective, and neglect the attribute of data distributed, it makes result appeared to uncertainty, this research hope propose a technology that have more effective and accurate, and could use it to be the fuzzy tool among numerous fuzzy time series methods.
According to the characteristics of data distributed, New method that variation and non- regular of interval to be proposed of this research, and with number of people entrance register university of Alabama as an example, the result is superior to other methods while compare with others that have used 4 different types data distributed, normal, uniform, poisson, exponential. This result has verified the prediction error of the method that this research institute proposes is far lower than the traditional fuzzy time series, the prediction ability to verify this method is the best at present.
This research so as to predict the technological method is applied to the stock price of Taiwan and predicted, it predicts the effect is better than the prediction of other methods too, the result of this real example proves the superiority of this research approach even more.
|
author2 |
Hai-Wen Lu |
author_facet |
Hai-Wen Lu Jing-Ru Tzeng 曾靖儒 |
author |
Jing-Ru Tzeng 曾靖儒 |
spellingShingle |
Jing-Ru Tzeng 曾靖儒 Applying Fuzzy time series for Stock Forecasting |
author_sort |
Jing-Ru Tzeng |
title |
Applying Fuzzy time series for Stock Forecasting |
title_short |
Applying Fuzzy time series for Stock Forecasting |
title_full |
Applying Fuzzy time series for Stock Forecasting |
title_fullStr |
Applying Fuzzy time series for Stock Forecasting |
title_full_unstemmed |
Applying Fuzzy time series for Stock Forecasting |
title_sort |
applying fuzzy time series for stock forecasting |
publishDate |
2005 |
url |
http://ndltd.ncl.edu.tw/handle/3589p2 |
work_keys_str_mv |
AT jingrutzeng applyingfuzzytimeseriesforstockforecasting AT céngjìngrú applyingfuzzytimeseriesforstockforecasting AT jingrutzeng yùnyòngmóhúshíjiānxùlièyúgǔjiàyùcè AT céngjìngrú yùnyòngmóhúshíjiānxùlièyúgǔjiàyùcè |
_version_ |
1719100824902172672 |