Algorithmic trading

using Random Forest

Classifier
 Definition
.
Algorithmic trading is a technique for executing orders
utilizing mechanized pre-modified trading guidelines
representing factors like time, cost, and volume

.
 Advantages of Algorithmic Trading
.
 Quick, Fast and Reduced Cost Trading

 Enhanced Precision and Diversity in Trading

 Backtesting enabling traders to assess and tweak a

trading idea.

.
 Random Forest Classifier
.
• Random Forest is a classifier that contains a number of
decision trees on various subsets of the given dataset and
takes the average to improve the predictive accuracy of that
dataset
• It is based on ensemble technique

.
 Algorithm
procedure RandomForestClassifier(D)
forest = new Array() .
for do i = 0 to B
Di = Bagging(D) . Bootstrap Aggregation
Ti = new DecisionTree()
featuresi = RandomFeatureSelection(Di)
Ti.train(Di,featuresi)
forest.add(Ti)
end for
return forest
end procedure

.
 Process .

Overview

.
 Out of Bag Error for Apple Stock

.
 Graph of OOB vs Number of Estimators

.
 Apple Stock Dataset Results

.
 Apple Stock Dataset Results

.
 Accuracy vs Different Machine learning models

.
 • Trade Size .

• Stop Loss
• Hedging

.
 References

• https://arxiv.org/abs/1605.00003
.

• https://www.sciencedirect.com/science/article/
abs/pii/S0957417413009731
• https://pdfs.semanticscholar.org/
6dd9/0b152dcb5cda46e3640e3648e46d28838372.
pdf

.
 Thanks!