Week 2 Blog

#

I worked mainly on cleaning/scraping datasets and analyzing them for phase 2 of the project. One website, in particular, required considerably more time to scrape compared to other data sources. The difficulty came from accessing different HTML text for each request, for which I had both professors help fix the issue. After that hurdle, scraping the rest of the website wasn’t too difficult, and I was able to acquire all of the datasets needed for the ML models.

I implemented a machine learning model for one of the visualizations, specifically to generate the line of best fit. With one of the functions I had to create in the homework, I successfully used a line of best fit into the graph. This allowed me to perform analysis and draw meaningful conclusions from the data. However, one thing to mention was that you could argue that graphs did not meet the assumption of linearity because there seems to be a oscillation pattern when comparing the stock price and date. However, we still thought that linear regression was appropriate because it met the other markov assumptions, and it is an extremely common practice to see linear regression being used to model stock prices over time in the financial world.

My thoughts: Honestly, today has been my favorite day in the dialogue so far. It was funny to see everyone’s studious side as they were working hard on the project but also being extremely goofy at the same time. As a business major, it is truly an honor to be apart of such an amazing, nerdy, and hilarious group of people.