1 Introduction

Over the past few weeks, we collected various forms of Twitter and stock performance data for a handful of technology companies. More specifically, we collected each company’s daily tweets directed at their respective support pages and their end-of-day stock price. Our data collection occurred over a span of 21 days, through the dates of 6/30/20 and 7/20/20. In our report, we utilized visualizations to show clear connections between social media and stock metrics, and ultimately showing relationships as a collective group.

1.1 Hypothesis

Before collecting and visualizing any data, in June, we predicted that we would see a negative correlation between each individual company’s stock performance and their respective Twitter support handle’s daily activity. We believe that this hypothesis represents that the better companies operate, with fewer errors in their product delivery, the happier and less active on social media support pages their customers will be. This heightened level of operational success and customer satisfaction or lack thereof is what we believe to directly affect each company’s stock price.

install.packages
## function (...) 
## .rs.callAs(name, hook, original, ...)
## <environment: 0x000002297da7b2c8>
library(httpuv)
library(rtweet)
library(dplyr)
library(lubridate)
library(ggplot2)
library(quantmod)

if (!requireNamespace("httpuv", quietly = TRUE)) {
  install.packages("httpuv")
}