Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
AI Stock-Screening Methodology for Portfolio Construction
Selecting profitable stocks is crucial in constructing an all-equity portfolio. Investors need to rely on screening mechanisms to aid investment decision making. New stock selection methods are highly desired, and existing methods are constantly improved. In this research, we investigate the potenti...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2021
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| Summary: | Selecting profitable stocks is crucial in constructing an all-equity portfolio. Investors need to rely on screening mechanisms to aid investment decision making. New stock selection methods are highly desired, and existing methods are constantly improved. In this research, we investigate the potential of relying on artificial intelligence to guide the stock selection process. The developed model employed genetic algorithms to optimize the selection of screening rules from among a set of widely accepted fundamental indicators. The model robustness and performance are tested using stock market real data over a 14-year period from 2006 till 2019. Based on portfolio quality factors of risk and return, the obtained results outperformed three commonly used stock screeners and the relative market indices as well. The findings of this work reveal that the proposed genetic algorithm provides a powerful dynamic tool to assist in screening and selecting valuable stocks.
JEL classification: G11, G17, C63
Keywords: Stock-Screening, Artificial Intelligence in Finance, Genetic Algorithms |
|---|