This is an attempt at algorithmic trading. I'll start off with a simple algorithm and expand from there.
Fortune 1000 company
- Drop of 5%+ in a single trading day
- No more than 40% off All Time high
- Have we purchased this stock in the last 1 day
- 30% return
- python
- poetry
- API keys for ALPACA and Twelvedata in environment variables.
- Uses TwelveData
- Specifically uses the price call. Example response:
{ "symbol": "AAPL", "name": "Apple Inc", "exchange": "NASDAQ", "mic_code": "XNAS", "currency": "USD", "datetime": "2021-09-16", "timestamp": 1631772000, "open": "148.44000", "high": "148.96840", "low": "147.22099", "close": "148.85001", "volume": "67903927", "previous_close": "149.09000", "change": "-0.23999", "percent_change": "-0.16097", "average_volume": "83571571", "rolling_1d_change": "123.123", "rolling_7d_change": "123.123", "rolling_period_change": "123.123" "is_market_open": false, "fifty_two_week": { "low": "103.10000", "high": "157.25999", "low_change": "45.75001", "high_change": "-8.40999", "low_change_percent": "44.37440", "high_change_percent": "-5.34782", "range": "103.099998 - 157.259995" }, "extended_change": "0.09", "extended_percent_change": "0.05", "extended_price": "125.22", "extended_timestamp": 1649845281 }
- Uses Alpaca