Abstract

The advent of digital age dramatically changed the way all aspects of commerce is conducted. From the largest multi-national conglomerates to the least small-and-medium enterprises and to the unassuming business-savvy individuals have adapted to take advantage of the benefits afforded by the resulting digital technology. Investing in, and profiting from, shares of stocks of companies listed in an organized stock exchange is one such instance. Gone are the days wherein stock investors and brokers are inseparable from their telephones to handle trades. Online platforms, powered by machine learning algorithms, have made investing in stocks not only accessible and convenient but also more lucrative. These online stock brokerage platforms account for the bulk of daily trades across the globe. Stocks investing, by its nature, is risky and, investors’ gains or losses are guided primarily by their risk appetite and ability to analyse movements in stock prices and the fundamentals of the company’s underlying financial information. The purpose of this study is to create a model that predicts or forecasts stock prices using machine learning algorithms. By designing a model that has predictive ability, investors are able to optimize gains or minimize losses. Stated differently, the availability of data on stock price forecasts allows investors to either buy, hold or sell stocks thereby enabling them to realize the highest possible gain or cut losses to the lowest possible level. This study aims to forecast stock prices for the next end-of-day trading of the top six publicly-listed Philippine conglomerates for the year 2019 as reported by Forbes. These conglomerates include:

- Ayala Corporation

- BDO Unibank Inc.

- SM Investments Corporation

-Top Frontier Investment Holdings (the controlling shareholder of San Miguel Corporation)

- Metropolitan Bank & Trust Corp

- JG Summit Holdings Inc.

Machine learning algorithms such as facebook prophet, random forest, auto regressive integrated moving average (ARIMA) will be used in this study to design a predictive model that generates the next end-of-day closing stock price with the highest accuracy.

Publication Date

12-2020

Document Type

Master's Project

Student Type

Graduate

Degree Name

Professional Studies (MS)

Department, Program, or Center

Graduate Programs & Research (Dubai)

Advisor

Sanjay Modak

Advisor/Committee Member

Ehsan Warriach

Campus

RIT Dubai

Share

COinS