Space Stock Predictor

• 

  GUI Interface

Inspiration

We are interested in using neural networks such as LSTMs to perform time series prediction to find the best times to buy/sell space-related securities.

What it does

Our app allows you to choose from three space ETFs (ARKX, UFO, and ROKT) and tells you when the best time to buy/sell in the near future is and the expected profit.

How we built it

We used Google Colab to create a train a LSTM model based on financial data pulled from Yahoo Finance of the three ETFs from 05/01/2022 to today.

Challenges we ran into

It was challenging to get the model to provide reasonable predictions at first.

Accomplishments that we're proud of

We're proud that our model now is capable of giving accurate predictions (based on backtesting predictions).

What we learned

Using on recent (~1 year) data, a LSTM model can provide accurate predictions for trading stocks.

What's next for Space Stock Predictor

We would like to add more securities for users to inquire about.

Built With

• colab
• python
• tensorflow