High-frequency time sequence volatility estimation method

Abstract

The invention discloses a high-frequency time sequence volatility estimation method. The method comprises the following steps: step 1, carrying out data preprocessing to obtain an original high-frequency yield sequence; step 2, removing the intraday periodicity to obtain a high-frequency yield sequence without the intraday periodicity; step 3, checking and removing the long memorability of the intraday periodic high-frequency yield sequence; step 4, establishing a high-frequency time sequence volatility model for introducing market microstructure noise; step 5, estimating model parameters of the volatility; and step 6, estimating to obtain the volatility of the high-frequency yield sequence without the intraday periodicity. The method overcomes defects that in prior art, interference of market microstructure noise on data is large, difference between obtained volatility and real volatility is large, and fluctuation situation of intraday level cannot be reflected. According to the intra-day volatility estimation method based on the high-frequency time sequence, estimation is more accurate, and a more accurate intra-day volatility estimation value can be obtained.

Description

technical field [0001] The invention belongs to the field of data modeling and analysis, and relates to a method for estimating intraday volatility based on high-frequency time series and considering the influence of market microstructure noise. Background technique [0002] The volatility of financial assets is one of the hotspots in the field of financial research, and is widely used in the fields of financial asset risk management and financial asset portfolio selection. Accurate measurement and accurate forecasting of volatility are crucial in the study of financial markets. In the volatility analysis of financial time series, volatility refers to the degree of dispersion of the return rate of financial assets from the mean value. In fact, real volatility is an unobservable latent variable. Therefore, how to construct a model that can not only describe the characteristics of volatility but also accurately estimate and predict volatility is extremely important. [0003...

AI Technical Summary

Problems solved by technology

[0017] In order to solve the above-mentioned technical problems, the present invention discloses a method for estimating the volatility of high-frequency time series. The present invention overcomes the large interference of market microstructure noise on data in the prior art, so that the obtained volatility is consistent with the real Volatility has a large difference and cannot reflect intraday fluctuations. The intraday volatility estimation method based on high-frequency time series of the present invention is more accurate, and can obtain more accurate intraday volatility predictions

Images