Each folder contains a project that utilizes several different computer science, financial, and statistical concepts to solve unique business problems. Below is a link to my Google Cloud that contains videos of the code I built and how some of these programs function in real-time:
https://drive.google.com/drive/folders/1DUw8DkCcvCtE4YSsbMG9Zf_vpRzjwl8w?usp=drive_link
Web Scrapping:
-
LinkedIn Applier is a script that automatically applys for jobs for a LinkedIn user. This program can apply to jobs that only require your contact infomration and resume. The limitation of this program is that it only applies for applications that do not require unique questioning.
-
Earnings Calendar Scrape goes the Nasdaq Earnings Calendar and filters all the companies that are releasing earnings reports after market close that day. This list is usually 50 stocks when it's "Earnings Season".
-
Nasdaq Historical Price retrieves the maximum daily historical prices for a given stock. This data is used for quantative stock analysis.
Regressions/Probabilities:
-
Real Estate Project utilizes a multi-linear regression model, backed with it's assumptions to predict Tax Assessed Property Value for a certain property in the Philadelphia, PA region. The project also includes a visual, descriptive analytics section that gives the user a better understanding of how the data is distributed and premliminary analysis on what variables may be significant in predicting Tax Assessed Property Value.
-
Earnings Model Project invovles the Naive Bayes Algorithm to predicts whether or not a specific, or multiple, stock(s) will increase within the range of 1% to 5%, the trading day following date that their earnings were released. The model justifies the variables used in the analysis, and is user friendly.
Natural Language Processing (NLP):
- Sneaker Project calculates the sentiment regarding a sneaker(s) that will be released in the near future. Understanding how the how people on Twitter feel about a certain shoe can be helpful to shoe buyers looking to resell. The project used Sentiment Analysis tools such as Co-sign Similarity, Topic-Modeling, and Python's Vader.