USING THE BLOOMBERG TERMINAL TO EVALUATE STOCK MARKET TRENDS: AN ANALYSIS OF HISTORICAL EPS FORECAST ACCURACY FOR ALL FIRMS IN THE 2019 S&P 500 INDEX

Marc Acebedo

Date of Award

2020

Document Type

Thesis

Degree Name

Bachelors

Department

Natural Sciences

First Advisor

Lepinksi, Matthew

Area of Concentration

Computer Science

Abstract

This thesis is divided into three sections documenting the process of cleaning, exploring, and interpreting the EPS and EOD data I gathered. The first section is data wrangling, where I explain how I gathered, assessed, and cleaned my data for quality and tidiness issues. During the assessment stage, I check for missing data (null values), duplicate data, and incongruous data types. Once satisfied with my clean CSVs, I stored them under the ./data/clean directory. The second section focuses on data exploration. This section is further separated into 3 stages: the univariate, bivariate, and multivariate exploration stages. I re-import the cleaned data, generate visualizations, make observations, and answer my 5 main research questions. Lastly, I focus on data explanation, where I narrow down my conclusions in a Jupyter Notebook Slides presentation. For each conclusion, I include all relevant visuals and graphs. This is the “storytelling” process of data analysis. All of the code for this project can be found on my Github repository at https://github.com/nihlan97/Evaluate-Historical-Stock-Market-Forecasts. The project is also hosted on my technical publication site, https://webbyboy.com/. To view the Jupyter Slides presentation of all my findings, download the reveal.js HTML file in my Github repository: data_explanatory.slides.html. Afterwards, render the file in the browser of your choice.

This document is currently not available here.

Share

COinS