Group 16 Project Proposal

1. Abstract

Stock is one of the common investment methods for HNWI (High Net Worth Investors). Traditionally, the visualization of different stocks is just line charts showing 
the price or the trend of the stock. However, with the development of visualization technologies, investors now may ask more for the stock graphs. This project 
tends to work out a visualization solution for stock investors by providing more analytical and interactive functions that allow users to see stock price 
forecasting, signals and payoffs of different trading strategies, and compare different stocks.

2. Objectives

The main objectives of the project are listed as follows:
• Demonstrate the trendline of different stocks in different industries and compare the difference between these stocks
• There are usually some rules hidden behind the graph and the project tends to uncover these rules of different stocks for investors to make better investment 
decision
• Make a diversified investment recommendations based on the similarities of the stocks and the preference of investors
• Forecast and visualize stock price using user specified forecasting method and time window.
• Visualize buy and sell signals on the stock graph, and compute the payoffs of customized trading strategies.
• Help users to analyze trading performance using various evaluation metrics.

3. Data source

• https://finance.yahoo.com/
• R package: quantmod, quantstrat

4. Data description

Basically, the data contains the price and the volume of a stock in a particular day. By using ‘quantmod’ package in R, users can specify the stock and timeframe 
that they want to look at.

5. Visualization tool used

• R: shiny, shinydashboard, ggplot2, plotly.
• Tableau
• Python: matplotlib

6. Challenges

• Data has large volume and complex pattern, thus might be difficult to model.
• The combinations of indicators can yield numerous trading strategies. Finding an optimal mix of different strategies is more of an art than a science.
• The stock price is vulnerable to many factors, thus analysis based on historical data may not give an accurate prediction for the future return.