Smart Stock Trading System

Project Plan
Version 1.0
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

Revision History
Date Version Description Author
20/10/2020 1.1 Detailed project plan for my project is
presented in this document

2
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

Table of Contents
1. Executive Summary 4

2. Introduction 5
2.1 Purpose 5
2.2 Scope 5
2.3 Definitions, Acronyms and Abbreviations 5

3. Project Overview 5
3.1 Project purpose, scope and objectives 5
3.1.1 Purpose 5
3.1.2 Scope 6
3.1.3 Objective 6
3.2 Assumptions and constraints 6
3.3 Project deliverables 6

4. Management Process 6
4.1 Project Cost Estimates 6
4.2 Project Schedule Plan 7
4.2.1 Project Scheduling 7
4.2.2 Key Milestones of the Project with Dates 8
4.3 Project Resources 9
4.3.1 Required Hardware 9
4.3.2 Required Software 9

3
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

Project Plan

1. Executive Summary
With the rapid development of information technology, Machine Learning based
automated stock trading systems has received more and more attention both from the industry
and academia. In this project we have planned to develop a smart stock trading system (SSTS)
that will make use of machine learning algorithms to analyze the patterns of stock prices and
automatically predicting the stock prices. This smart stock trading will assist in decision-making
and in the prediction of key variables such as stock prices and earnings. This system will be able
to quickly process a large amount of historical stock prices data that affects stock market price
fluctuations in order to fundamentally improve the accuracy and efficiency of forecasting.

We will use data comprise of Brent Crude Oil continuous futures since 2000 for training
of our machine learning model and this dataset consists of 11 variables. These individual futures
contracts trade for a very small time period, and hence they are not suitable for long-horizon
analysis. Continuous futures contracts resolve this issue by chaining together the series of
individual futures contracts, for providing a long-term history of prices which will be suitable for
trading, behavioral and strategy analysis. Before performing the analysis, necessary data cleaning
and data wrangling will be performed by transforming the data and mapping it into a format that
will my more suitable for performing downstream data analytics. When applying the Machine
Learning to stock data, we will also perform technical analysis to see if our algorithm can
accurately learn the underlying patterns in the stock time series. Before perform collinearity
statistics necessary cleaning was will be performed. We will be used selected features which will
be based either on their stand-alone characteristics or on their relationship with the target.

The accuracy of the developed machine learning algorithm will be tested by checking the
predicted stock prices values with the actual prices. In this way will be able to test if our smart
stock trading system and its underlying machine learning algorithm is working accurately or not.

4
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

2. Introduction
In this document I have presented the project plan of machine learning based Smart Stock
Trading System. In this project I am going to develop a machine learning based Smart Trading
System that will analyze the price patterns and will predict the stock prices.

2.1 Purpose
The project plan will serve as the basis that how we will be planning the different project
activities and how we will control the overall progress of the project to achieve the objectives of
the software project. This document will help to plan the best use of people, time, money,
resources in the whole project life cycle. This project plan provide evidence that the major
development related issues have been well thought out in advance by the development team.

2.2 Scope
This document will describe the overall plan of the project that how will perform
different project tasks to achieve the project goals. The project overview section discusses the
purpose of the project and its scope, project related constraints and the deliverables of the
project. Project management process covers the project cost and schedule estimates. Resources
needed to carry the project activities and different risk involved in the development of the
project.

2.3 Definitions, Acronyms and Abbreviations

ML – Machine Learning

SRS – System Requirement Specifications

SDD – Software Design Document

3. Project Overview

3.1 Project purpose, scope and objectives

3.1.1 Purpose
The purpose of this project is to develop a machine learning based Smart Trading System
that will analyze the price patterns and will predict the stock prices. In this project we will be

5
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

working on the development of automated stock prediction and recommendation system that will
make use of machine learning algorithm for predicting stock prices.

3.1.2 Scope
This smart trading system will assist in analyzing price patterns and will help to predict
stock prices based on different situations and conditions. A machine learning algorithm will be
developed and trained with historical data of stock prices and then based on that training this
algorithm will be able to predict stock prices. I will be working on a classification machine
learning algorithm with the assistance of feature selection to predict that what will be the success
rate of making profit.

3.1.3 Objective
The objective of this project to develop a machine learning algorithm for stock trading.

3.2 Assumptions and constraints

 Python should be used for the coding of this project.

 The accuracy of the machine learning algorithm that will be developed in this project
should be not less than 90%.

 Limited funding sources are available and there is no contingency.

3.3 Project deliverables

 Technical documentation

 User documentation: user manuals, installation guide

 Software and dataset for training of machine learning model

4. Management Process

4.1 Project Cost Estimates

The estimated cost for the development of this project is $21560. The detailed
estimation of the project cost is given in the table below:

Unit/Hr Cost/Unit/H Sub Level % of Total

6
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

s. rs. totals Total

1. Project $3500 16.2%
Management
Project Manger 80 $25 $2000
Team Members 100 $15 $1500
2. Requirement $1260 5.8%
Engineering
Licensed Tools $60
Team Members 80 $200 $1200
2. Hardware $800 $800 3.7%
3. Development $6100 28.3%
Licensed Software $100
Team Members 300 $20 $6000
4. Testing $4000 $4000 18.6%
5. Training and $1900 8.8%
Support
Trainee Cost 70 $20 $1400
Travel Cost 50 $10 $500
6. Deployment Cost $1500 7.0%
Software Deployment $1500
7. Reserves $2500 $2500 11.6%
Total $21580

4.2 Project Schedule Plan

4.2.1 Project Scheduling

Here is the schedule of the project

No. Activity Duration(days) Resources

3 Requirement Elicitation 05 Microsoft Word
4 Requirement Analysis 04 Requirement
Analysis Tools
5 Requirement 04 Microsoft Word
Specification
6 System Design 06 MS Visio

7
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

7 Implementation and Unit 42 Jupyter Notebook,

Testing Dataset, Test Cases,
Test Plan,
8 Integration and System 08 Test Cases, Test
Testing Plan
9 Deployment 02 Personnel
Computer

Figure 1: Gannt Chart of the Project

4.2.2 Key Milestones of the Project with Dates

Key Milestones of the Project with dates

S. No Elapsed time since start Milestone Deliverable

of the project
1. Requirement Requirement Draft,
Elicitation and Requirement Analysis
Analysis Document
2. Requirement Software Requirement
Specification Specification (SRS)
3. System Design Software Design Document
4. Implementation and
Ready to Deliver Product
and Testing

8
Smart Stock Trading System Version: 1.0
Project Plan Date: 27/10/2020

5. Deployment Algorithm, Software System,

Dataset

4.3 Project Resources

4.3.1 Required Hardware

 Laptop (Core i5, 8GB RAM, 2.00 Ghz x 2 Cores CPU)

4.3.2 Required Software

 Jupyter Notebook

 MS Visio

 MS Word

 MS Excel

 Python 3.0

 Python Libraries (Numpy, Scipy, Scikit-learn, TensorFlow, Pandas, Matplotlib)