padibona / FinTech-Homework-14

LSTM Stock Predictor

In this assignment, I used deep learning recurrent neural networks to model bitcoin closing prices. One model used the FNG (Fear and Greed) indicators to predict the closing price, while the second model used a window of closing prices to predict the nth closing price.

I will then compare the 2 models and answer the following questions:

• Which model has a lower loss?

• Which model tracks the actual values better over time?

• Which window size works best for the model?

Model 1 - Using FNG indicator.

FNG Notebook

• optimal window_size = 1
• loss: 0.1165

Plot of FNG model vs actual:

Model 2 - Using closing prices.

Closing Prices Notebook

• optimal window_size = 4
• loss: 0.0055

Answers to questions:

• Which model has a lower loss?
  • Most certainly the 'Closing Price' model has a lower loss.
• Which model tracks the actual values better over time?
  • Also, the 'Closing Price' model tracks the values better over time.
• Which window size works best for the model?
  • For the FNG model, the best window size was 1.
  • For the Closing Price model, the best window size was 4.