nanoback 1.0.1

C++20 event-driven multi-asset backtesting engine with paper-trading bridge, analytics dashboard, WFO/Monte Carlo validation, and execution-realistic simulation.

BACKTESTER (nanoback)

nanoback is a C++20 event-driven backtesting engine with Python APIs for research, paper trading, and production-grade validation workflows.

What's New (v1.0)

• Live bridge (v0.6): PaperBroker streams real-time ticks into the same engine path as backtests
• Safety reconciliation: Reconciler detects and optionally auto-corrects engine-vs-broker position drift
• Interactive reporting:
  • result.plot() equity + underwater drawdown
  • result.to_html("report.html") self-contained shareable report
  • result.dashboard() Streamlit analytics panel
  • sweep_result.heatmap(...) Plotly parameter robustness diagnostics
  • wfo_result.plot() IS vs OOS Sharpe + OOS equity
• Release quality:
  • Multi-platform wheel publishing through cibuildwheel
  • Benchmark regression guard on CI merges to main
  • Expanded edge-case regression suite

Install

pip install nanoback

Optional extras:

pip install "nanoback[io,viz]"

Core Example

import numpy as np
import nanoback as nb

result = nb.run_backtest(
    timestamps=np.array([1, 2, 3, 4], dtype=np.int64),
    prices=np.array([100.0, 101.0, 99.0, 102.0], dtype=np.float64),
    signals=np.array([1, 1, 0, -1], dtype=np.int64),
    config=nb.BacktestConfig(max_position=2),
)

print(result)
fig = result.plot()
result.to_html("report.html")

Paper Trading Bridge (v0.6)

from datetime import datetime, timedelta, timezone
from nanoback.paper import PaperBroker, PaperTick
import nanoback as nb

class Feed:
    def __init__(self, ticks):
        self._ticks = list(ticks)
    def next_tick(self, timeout_seconds=None):
        return self._ticks.pop(0) if self._ticks else None
    def fetch_positions(self):
        return {"AAA": 0.0}
    def submit_order(self, symbol, quantity_delta):
        pass

def strategy(tick, state):
    return {"AAA": 1}

broker = PaperBroker(
    symbols=["AAA"],
    strategy=strategy,
    feed=Feed([PaperTick(timestamp_ns=1, symbol="AAA", price=100.0)]),
    config=nb.BacktestConfig(max_position=2),
)
broker.run_until(datetime.now(timezone.utc) + timedelta(seconds=1))

Repo Layout

• cpp/, include/nanoback/: core engine and bindings
• python/nanoback/: analytics, live bridge, strategy, sweep/WFO/MC modules
• benchmarks/: latency/stress harnesses and baselines
• tests/: regression suite

Development

python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install -e .[dev]
python -m pytest