Study on the Price-Volume Relationship with The Genetic programming Master Curve for price-impact function

• Main Authors: Hui-Wen Lin, 林慧紋
• Other Authors: 李沃牆
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/89527331791825210889

碩士 === 淡江大學 === 財務金融學系碩士在職專班 === 102 === This study analyzes price and volume relations of top ten industrial sub-indices in the Taiwan Stock Exchange Index through the correlation coefficient, and then to identify the price of top ten industrial sub-indices impact function (or a function of the relationship between price and volume) by genetic programming, and adding market capitalization, exchange rates, interest rates to explore the top ten industrial sub-indices price correlation use of regression analysis. Conclusions obtained are as follows: (1) the correlation coefficient analysis results, the top ten stock index in each time interval part presents a negative relationship, but the top ten industrial sub-indices in the whole period were highly positive relationship between price and volume relationship, the long-term trend showing that a large return is usually accompanied by a large trading volume. (2) find out the price impact function through genetic programming, and found that within sample the minimum mean square error of top ten industrial sub-indices is Electric Machinery, in out-sample the minimum mean square error is the Iron and Steel, which means that Iron and Steel has the best prediction. (3) Regression analysis found that the market capitalization change rate and the price movements of top ten industrial sub-indices were highly positive relationship, between exchange rate changes and price changes in addition to food and Iron and Steel relationship is not significant, the other eight industrial sub-indices are negative, the relationship between interest rates and the prices of the top ten industrial sub-indices are less significant relationship.