Outlier Detection And Bias Correction In Financial Time Series And Investment Consulting Analysis

The aim of this paper is to get the outlier disturbance amplitude approximate unbiased esti-mation of time series model, in order to improve the effect of diagnosis model, at the same time can be the approximate unbiased estimation applied to need accurate calculation of outlier cases.This paper mainly researches ARMA model and ARCH model.As linkage benefit is increasingly frequent in the economic field,the use of time series model is a very important method in the description and prediction of the objective economic process. However, in practical applications, due to the special economic areas, using traditional statistical methods in economic time series model analysis often encounter many difficulties. Since ARCH model has great benefits in financial consulting, this paper also use the actual data in investment in consulting company to correct the outlier, which makes the logic of the report more reasonable.