Research on the Augmented Dickey-Fuller Test for Predicting Stock Prices and Returns

Abstract

With the continuous accumulation of theoretical knowledge and progressive applied research, analyzing financial time series data gradually becomes everlasting research in modern days. The simpler Dickey-Fuller originally is a test commonly used in econo-metrics and finance to test the stationarity of financial time series data. Thereafter, simpler Dickey-Fuller is eventually extended to the augmented Dickey-Fuller test to examine the stationarity of financial time series data such as stock prices, returns, and so on. This paper mainly focuses on the utilization of the augment Dickey-Fuller test and tests the stationarity of stock prices and returns for Nike, and Amazon. Both the stock price which is non-stationary, and the return, which is stationary, illustrate that these two companies are market-efficient. Additionally, the paper provides plots of stock prices and returns for these two companies by executing Python code. The results from the augment Dickey-Fuller test not only verify the characteristics of these plots but also indicate that the augment Dickey-Fuller test is useful to predict stock price and return.