Implement New Indicator

Implement New Indicator

Scope

Use this skill when the user asks to implement a new strategy indicator in TradePy.

Core rules from the project:

• Implementations go into tradepy/strategy/indicators.py
• Use _fast_ewm for exponential weighted averages. We deliberately avoid Wilder's Smoothing due to its complexity, and instead use Polars' built-in ewm_mean for best performance. As a result, more leading bars need to be discarded during warmup.
• For each new indicator implemented, verify it against TA-Lib. The dev dependency already includes TA-Lib. You need to confirm that the calculated indicator values converge with TA-Lib's results.

Implementation Workflow

1. Read tradepy/strategy/indicators.py, tradepy/strategy/__init__.py, and relevant tests before editing.
2. Add the indicator as a typed @dataclass(frozen=True) subclass of SeriesIndicator and implement compute(self, value: pl.Expr). The base class resolves the input: at the pipeline root, value is the adjusted column price; otherwise it is the upstream pipeline output.
3. Follow the existing output style:
   • Return pl.Expr for single-output indicators.
   • Return dict[str, pl.Expr] for multi-output indicators that require Take(...).
   • Set requires_upstream: ClassVar[bool] = True for transforms that are meaningless without a piped input (e.g. Lag).
4. For exponential weighted averages, use _fast_ewm(...) and set a warmup long enough for convergence against TA-Lib. Declare the warmup multiplier as a WARMUP_FACTOR: ClassVar[int] on the indicator class.
5. Export the indicator from tradepy/strategy/__init__.py.
6. Add focused pytest coverage in tests/test_indicators.py.

TA-Lib Verification

Tests must compare the new Polars indicator against the matching TA-Lib function and prove convergence after warmup.

Use a deterministic price series with enough rows for convergence, then compare only the non-null tail:

import numpy as np
import talib


def test_new_indicator_converges_with_talib() -> None:
    close = np.linspace(10.0, 200.0, 300)
    # Build a Polars DataFrame and compute the project indicator.
    # Compute the TA-Lib reference with the same period/configuration.
    # Drop leading null/NaN values and assert the tail converges.

Prefer pytest.approx(...) or numpy.testing.assert_allclose(...) with tolerances that reflect floating-point convergence rather than exact equality.

Commands

Use uv for verification:

uv run pytest tests/test_indicators.py

If type or lint behavior is relevant to the change, also run the existing project commands for pyright or ruff.

Source: namoshizun/TradePy — distributed by TomeVault.