Abstract: Stock forecasting is challenging because of stock volatility
and dependability on external factors, such as economic, social, and
political factors. This motivates investors to seek tools to identify
stock trends to reap profits.
In this research, we compared several heterogeneous ensembles for
financial forecasting, including averaging, weighted, stacking, and
blending ensembles. In addition, we used a random forest regressor as
Regression was used to predict the next day’s closing stock price. We
used classification to label closing stock value as HIGH or LOW by
comparing with the opening stock value of a particular company. We used
Long Short Term Memory (LSTM) models, Linear Regression, and Support
Vector Machines (SVM) as individual models. Further, we analyzed 10
years of historical data of the most active 20 companies of the NASDAQ
stock exchange for implementing ensemble models.
In conclusion, experimental results depict blending ensembles perform
the best out of compared ensembles in financial forecasting. Further,
they reveal SVM is under-performing, LSTM outputs are satisfactory,
while linear regression produced promising results.
Data: Data for this research was gathered from online available sources
from the NASDAQ American stock exchange.
We gathered data for most active 20 companies and 10 years of historical
data from 21st September 2019 backwards. We used 40044 data points in