Over the course of the project I investigated what kinds of inputs we should use for the machine learning models. I also did an analysis on the inputs to see if they are statistically significant or random. I did this by using multi-variable linear regression in R and looking at the P-values. I was also in charge of explaining all the economic indicators, where we got them, how we transformed the data, and how these impact our independent variable. I displayed our linear results in a dashboard in tableau and imposed that over inflation data to show how the two were related.

The greatest personal challenge over the project was which machine learning model were we going to use in order to accurately predict the SPY price. This was challenging as my understanding of machine learning was very limited. Luckily my teammates and the data I was able to find on Kaggle led us to doing a neural network. This was still challenging as the code was very dense and technical to understand.

The team used slack, zoom, google slides, and google sheets to communicate. We were initially challenged by how we were going to visual everything from Python and how we were going to input it into tableau. We eventually resolved this by exporting everything to csvs and having each data point match up with pre-existing data by using a number of Vlookups. Next time we would like to have more economic indicators and we would like to use web scraping to see market sentiment of the stock market to hep predict price. Our strengths were that we were all familiar with and had an interest with the topic that we chose. We were more easily able to talk about and give our input on the topic just based on our intuition. We were overall interested in seeing if we are able to accurately predict the price of the SP&500 though machine learning. At the end of it all on the day of our presentation our neural network was able to predict the price with $10 of the SPY price and overall we are pleased with how the model was able to predict the price up to 2 years out.