Stock Market Modeling and Forecasting

Market turning points forecasting using wavelet analysis

“…425 Based on the system adaptation framework we previously proposed [26,27], its internal model is used in this paper to capture the dynamical properties of stock markets and generate a signal-rich residue for turning points forecasting. The MRA of the DWT and MODWT are used to decompose the internal residue and further extract its middle-frequency signals.…”

“…One of the advantages of this system adaptation framework is its structure, within which not only quantitative relationships between the market and external influences can be modeled, but also the market dynamics and behaviors are well captured. Our study found that the residue of the internal model contains predictive signals on the market cycles [27]. This 80 paper first applied the internal model to the stock prices to generate a signalrich residue series.…”

“…The input r is consisted of external influences and the outputp is the estimated security price. The actual security price p is In this paper, we focus on the internal model, and for the readers who are interested in the detail of our system adaptation models or the design of adaptive filter, please refer to our previous publications [26,27]. To capture the slow dynamic properties, the internal model works as a price trend generator, 105 which makes the estimated prices have the same trends as the actual prices.…”

“…There are some previous works using Fourier methods to study the market turning periods [27,28]. However, Fourier transform assumes the signal is periodic.…”

“…The model is based on our previous system adaptation framework [26,27], in which the financial market is considered to be a complex 75 dynamic system, see Figure 1. One of the advantages of this system adaptation framework is its structure, within which not only quantitative relationships between the market and external influences can be modeled, but also the market dynamics and behaviors are well captured.…”