Trend model estimation of stock time series based on outliers

To study the application of outliers in stock market, here a method detecting the outlier of time series of stock prices which is based on d-nearest neighbor clustering is presented. Study on the outlier of time series in stock market is dealt with in three stages, in the first stage, the breaking points where the short-term trend changes are used to be observing points. And then, the d-nearest neighbor clustering method is introduced to detect the outliers. In the third stage, the short-term trend of outliers is used for linear trend modeling with the least square method forecasting the future trend of stock time series. Empirical study, based on the historical data of Shanghai stock exchange and Shenzhen stock exchange, proved that the method is feasible and effective.