Semantic analysis on wallstreetbets using VADER and Mixed-Effect model.

This project was my bachelor thesis. It was an idea I developed by myself following the big fuzz around Reddit’s wallstreetbets and their opposition towards a short position regarding the stock GME. I conducted semantic analysis on the comments from the most discussed threads using VADER for a period of 15 days. I also retrieved information about the traded volume of the mentioned stocks over the same period of time.

Thereafter I analyzed the result using a Mixed-Effect model, which is a type of multiple linear regression model which also takes into considerations differences between clusters.

I am very pleased with how well this project went and I dare to say that so was my examinators. If you want to read more about the code I used etc, please visit my github library S-A_Vader.

I created this website with Hugo.

I wanted to broaden my expertise and seemingly this was a great way to produce something usefull as well as making me more aquainted with:

• Github

• Static websites

• Markdown

• Servers and Domains