Testing the Predictive Power of Machine Learning Algorithms for Stock Market Movements Based on Air Pollution Data

Kelvin Lee Yong Ming ()
Additional contact information
Kelvin Lee Yong Ming: Taylor’s University

A chapter in Industry Forward and Technology Transformation in Business and Entrepreneurship, 2023, pp 151-160 from Springer

Abstract: Abstract Air pollution has seriously threatened the lives of mankind. Governments throughout the world are taking several steps to reduce the impact of air pollution. Several recent studies found that variations in air pollution adversely affect the stock market movement by using the conventional statistical model, such as the fixed effect model and quantile regression. This study attempts to narrow down the methodological gap by testing the predictive power of machine learning algorithms for Singapore stock market movements based on air pollution data. Specifically, this study tested five machine learning algorithms—(i) Random Forest, (ii) XGBoost, (iii) ADaBoost, (iv) Support Vector Machine, and (v) K-Nearest Neighbour. The input data for the prediction comprised the closing prices, and index for PM 2.5 and PM 10. The accuracy of prediction was further measured by using MAE, MAPE, MSE, and RMSE. The results indicated that XGBoost has the highest accuracy in predicting Singapore’s stock price movements. The findings also suggest that the 1 day average (value from the previous day) of the closing price, and the index for PM2.5 and PM10 are suitable for the prediction of stock market movements. These findings serve as a guideline for stock market prediction among market participants when considering air pollution.

Keywords: Stock Market; Machine Learning; Air Pollution; Digital Technology (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-981-99-2337-3_14

Ordering information: This item can be ordered from
http://www.springer.com/9789819923373

DOI: 10.1007/978-981-99-2337-3_14

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().