A Study of Stock Market Predictability Based on Financial Time Series Models

Abstract

In today’s era of economic globalization and financial integration, the stock market is constantly complex, showing many deviations that cannot be explained by classical financial analysis, but at the same time, some classic financial statistical features have striking similarities. This suggests that although the stock market is intricate, there are universal laws that can be found through data mining to find its underlying operating rules. In this paper, we construct financial time series models such as ARIMA, ARCH, and GARCH to predict the stock market price fluctuations and trends. The ARIMA model is used to fit the linear financial time series, and the GARCH model is used to fit the nonlinear time series residuals. The results show that the integrated tree model based on the idea of weight voting has high accuracy in predicting stock market bulls and bears, with XGBoost prediction accuracy up to 96%, and the neural network model is also very effective, with an accuracy rate of over 90%.

MAGRU: Multi-layer Attention with GRU for Logistics Warehousing Demand Prediction

“…Time series forecasting is the basis of warehouse demand forecasting and has remained a widely studied hot issue. Classical time series forecasting methods are ARIMA, ARCH, and GARCH [6][7]. However, there is a limitation that only linear models can be built, while realistic prediction tasks contain linear features and are often influenced by nonlinear features.…”

“…An encoder is essentially an RNN. For the time series prediction problem, given an input sequence, the encoder is used to learn the mapping from x to ht as shown in (6), where t represents the time step.…”