A Study on the Effectiveness of SPY ETF Prediction Based on Signal Trading
<p>Yiwen Liu</p> · Academic Journal of Business & Management · 2024
An exchange-traded fund (ETF) is a popular trading tool in today's stock market, known for its low cost, low risk, and high liquidity. Enhancing the accuracy of predicting ETF future prices using quality trading strategies is of interest. This study selects the widely traded SPY ETF in the U.S. market as a representative. It utilizes market signal theory and panel data of daily closing prices of five major international stock indices from 2012 to 2021. Python Jupiter notebook is used as the programming tool to construct a multiple regression model and conduct a series of predictive analyses. The aim is to track SPY price changes by measuring the fluctuations in other stock indices, thereby predicting SPY prices. Following regression predictions, a time series prediction model is employed, and the market signal predictions are compared with a Buy-and-hold strategy. Lastly, using metrics like R2, Adj-R2, Sharpe ratio, and Maximum drawdown as evaluation criteria, empirical results are observed. The findings indicate that the Signal-based trading strategy outperforms Buy-and-hold, yet the impact of several stock index prices on ETF regression is not significant. Investors should adhere to cautionary principles, consider market signals, and diversify investments to mitigate risks.