This project was created during my summer internship at lemon.markets
A berlin-based Fintech startup which is developing a Brokerage API and offering a Trading and Market Data API
This is a public lemon.markets repository that showcases how to utilize Facebook's Prophet model in an attempt to forecast the close price of the Tesla Stock. The repo consists of two folders. TSLA - c prices contains the model built using the actual closing prices while TSLA - c differences contains 3 models built using the price difference, price log difference and price percentage difference.
You can find an article linked to this repo here
- sign up to lemon.markets
- clone this repo
- Change the market_data_api_key in
get_data.ipynbwith your personal lemon.markets API key - run
get_data.ipynbto get the csv file containing the stock data - run
prophet-model-v3-Hyperparameter tuning.ipynbto evaluate the best values for your parameters - feed the parameters from step 5 into your model in
tuned_model.ipynb - forecast away
This project uses the lemon.markets API and the lemon.markets Python SDK.
🍋 lemon.markets is a brokerage API by developers for developers that allows you to build your own experience at the stock market. We will use the Market Data API and Trading API to build a mean reversion trading strategy in this project!
If you do not have a lemon.markets account yet, you can sign up at lemon.markets.
lemon.markets is an API from developers for developers and this (and all lemon.markets open source projects) is(are) a work in progress. Therefore, we highly encourage you to get involved by opening a PR or contacting us directly via support@lemon.markets.
Looking forward to building lemon.markets with you 🍋