Forecasting Covered Call Exchange-Traded Funds (ETFs) Using Time Series, Machine Learning, and Deep Learning Models
This study explores the application of time series, machine learning (ML), and deep learning (DL) models to predict the prices and performance of covered call ETFs. Utilizing historical data from major covered call ETFs like QYLD, XYLD, JEPI, JEPQ, and RYLD, the research assesses the predictive accuracy and reliability of different forecasting approaches. It compares traditional time series methods, including ARIMA and Heterogeneous Autoregressive Model (HAR), with advanced ML techniques such as Random Forests (RF) and Support Vector Regression (SVR), as well as DL models like Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN). Model performance is evaluated using metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). Results indicate that the DL models are effective at identifying the nonlinear patterns and temporal dependencies in the price movements of covered call ETFs, outperforming both traditional time series and ML techniques. These findings enhance the existing financial forecasting literature and offer valuable insights for investors and portfolio managers aiming to improve their strategies using covered call ETFs.