The inspiration for Pecunia Ex Machina (PEM) by our team's love of technology and money, in addition to our awareness of the barriers to stock trading for the average person. While the majority of Wall Street trades are made by computers, this technology is kept out of the hands of everyday investors, we aim to change this.
What it does
PEM utilizes a neural network in order to identify behavioral patterns in individual stocks, predicting microeconomic trends in order to advise the user to either buy or sell shares in said stock.
How we built it
We developed PEM's neural net from scratch, designing it to train itself on the past year's data to recognize the behavioral trends of a stock, and then using the past 10 days of data to predict the future value of the stock to make a recommendation to buy or sell.
Challenges we ran into
The biggest challenge we ran into while developing PEM was definitely the UX/UI development as our team's development experience in the area is limited.
Accomplishments that we're proud of
We are especially proud of the neural network and its precision in determining future value.
What we learned
We gained substantial experience in neural network development as it was made from scratch, giving us keen insight into the intricacies of their development.
What's next for Pecunia Ex Machina
We hope to further develop the neural net to move toward deep learning in effort to significantly improve the prediction accuracy. We hope to implement functionality that will allow PEM to make quantitative predictions of future market data such as opens, closes, highs, and lows.