dwker/strategy-lab
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

notes added

Browse Source
This commit is contained in:
David Brazda
2024-08-31 06:58:49 +02:00
parent cce5bbdb0b
commit 602e6aae9f
1 changed files with 10 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
README.md
View File

@ -1,6 +1,6 @@
# Note for The Team
Entry points are in `strat_STRATEGYNAME` directories.
Entry points are in `strat_STRATEGYNAME` directories in `research`.
there are `SINGLE` and `MULTI` version
- `SINGLE` - just straight one pass meaning fetching data (local/remote), indicators and entry-points and backtest with SINGLE parameter values and strategy result. Used for basic strategy research.
@ -8,6 +8,15 @@ there are `SINGLE` and `MULTI` version
I can imagine it can be used to create MVP (not use with notebooks as they become clumsy with lot of data), hyperparameter testing should be refactored, walk forward optimization and cross validation should be added.
Maybe start exploration with `/research/strat_ORDER_IMBALANCE/v2_single.ipynb`
strategies I have worked on so far
- CANDLE_GAPS - candle gaps of various size within high resolution ohlcv (1s) as source for entry signals
- LINREG_MULTI - combination of linear regressions on different length and time resolutions
- SUPERTREND - supetrend on different resolutions
- ORDER_IMBALANCE - order imbalanced columns calculated from trade data as new columns on OHLCV bars
- TIME_ENTRIES - just idea for time based entries to leverage random movement of price
# Research for v2realbot
## Overview