The study of building the price prediction model for electronic component retailing market

• Main Authors: Hsiu-Fen Cheng, 鄭秀芬
• Other Authors: Yin-Yi Chang
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/58476245077015375573

碩士 === 輔仁大學 === 資訊管理學系 === 91 === The Price trend of stock is always a critical factor when retailers need to replenish the stock of merchandise. It will influence how many products need to be replenished and what price should be made. Because there are price variations of electronic component in retailing market, making a price is used as a weapon in market competition. In lately years, prediction technology is getting mature and used in prediction of society phenomenon. In this research, we collect history retailing prices of three different electronic components：main board、graphic card and LCD monitor ,and research the feasibility of grey prediction model and neuron network in electronic components price prediction of retailing market. We find the main board prediction result of the grey prediction model is better than the result of neuron network. There are no obvious difference between prediction result of grey prediction model and neuron network in graphic cards, but building a grey prediction model is easier than building a neuron network prediction model. Such that, we suggest to use grey prediction model as graphic cards’ prediction model. And we also find out that the prediction error of neuron network is smaller than grey prediction model’s prediction error in LCD monitor’s prediction, it shows neuron network is more suitable in prediction of LCD monitor than grey prediction model.