Abstract

In this research, I investigate the performance of three technical trading strategies—Moving Average Crossover, Bollinger Bands, and a combined Bollinger Bands with RSI ap- proach—applied to three major U.S. stocks: Apple (AAPL), Microsoft (MSFT) and Meta (META). By systematically tuning key parameters and conducting in-sample and out- of-sample testing, the study aims to evaluate the effectiveness and robustness of these strategies. The results show that it is possible to identify parameter sets that outperform the SPY benchmark during the in-sample period. In particular, the combined Bollinger Bands and RSI strategy achieved the highest returns on META. However, out-of-sample analysis reveals that Method 3 did not execute trades, highlighting limitations in signal adaptability. Meanwhile, Methods 1 and 2 maintained stronger performance on MSFT and META during the out-of-sample period but failed to beat the benchmark on AAPL. These findings emphasize the trade-offs between optimizing historical performance and ensuring future robustness. In general, this study underscores the importance of combining technical signals with adaptive parameter management to improve the resilience of strategies in real world markets.

Advisor

Long, Colby

Department

Statistical and Data Sciences

Disciplines

Data Science

Publication Date

2025

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Share

COinS
 

© Copyright 2025 Bat-Orgil Sainbuyan