Can We See the Beats of Financial Markets: Studies on Stock Trading Activities

Can We See the Beats of Financial Markets: Studies on Stock Trading Activities
By:
Published on 2005 by ProQuest

The purpose of my research is to investigate (1) the dynamic patterns of trading activities on the stock market across different time scales ranging from low frequency such as daily, weekly and monthly to high frequency such as tick-by-tick; (2) the relationships of the dynamics among different time scales. This thesis consists of three chapters; (3) the cross sectional studies of these dynamics across different time scales. Chapter one studies the dynamics of daily stock trading volume by using the CRSP data starting from June 1962 to December 2002. I used a local polynomial smoothing combined with a local Whilttle estimation method to investigate its dynamics. Its major result is that daily stock trading volume is a long memory process, which implies that current trading volume has a substantial impact on the trading volumes in even more than one year. In addition, it implies that trading volume is a wild stochastic process defined in Mandelbrot (1997). Chapter two studies the stock trading time by using the Trade and Quote data from 1993 to 2002 complied by New York Stock Exchange. The trading time is defined as the number of trades at particular time. Using wavelet as a major analysis tool, I found that the trading time for all stocks investigated in this chapter has qualitatively similar patterns: (1) its intra-day dynamics follow non power laws in time scales; (2) its inter-day dynamics follow power laws in time scales. This result explains the daily dynamics of the trading volume as a long memory process and also establishes links between intra day trades with inter day trades. Chapter three develops a multivariate time series model to detect the common movement across stock trading volumes. The common movement of stock trading volumes is defined as a fractional cointegration among stock trading volumes, which is a long run relationship among them. By using the daily stock trading volumes of DJ30 components, I found several delicate long run comovement among them.

This Book was ranked 32 by Google Books for keyword stock trading.