AI-forexPredict

A library aimed at providing utilities for computing indicators data and also building a prediction ML model and other

Dependencies

The library is developed with the help of gradle tool and JDKversion 18

dependencies {
    testImplementation(platform("org.junit:junit-bom:5.9.1"))
    testImplementation("org.junit.jupiter:junit-jupiter")
    implementation("org.glassfish.tyrus.bundles:tyrus-standalone-client:1.9")
    implementation("org.json:json:20230227")
    implementation("com.aparapi:aparapi:3.0.0")
}

Packages included so far.

Data Collection and Pre-processing

For collection and cleaning of forex data and technical indicators from various sources. In here there is usage of a default API to fetch forex or instrument data

Feature Engineering and Selection

For generating and selecting event-driven features indicating a change of trend in direction. This involves implementation of some of the technical indicators relevant for trend prediction among which include; Trend, Oscillators and Volume indicators.

Technical Indicators

This is partial development and so does not contain all indicators as the other has be added in the next cycle of development

Trend

• Exponential Moving Average
• Simple Moving Average
• Smoothed Moving Average
• Linear Weighted Moving Average
• Standard Deviation

Oscillator

• Relative Strength Index
• Commodity Channel Index
• Average True Index
• Stochastic Oscillator

ML Models

Support Vector Machine (SVM)

• CPU - Implemented to run on the CPU
• GPU - Implemented and capable of training on GPU

Long Short-Term(LSTM) Memory

• CPU - Implemented to run on the CPU
• GPU - Implemented and capable of training on GPU