Donchian breakout

easyTradingacceptance —~15 min

Implement donchian_signals(highs, lows, closes, window) — for every bar t >= window, emit 1 if the close is STRICTLY above the highest high of the previous window bars (highs[t - window : t], today excluded), -1 if STRICTLY below the lowest low of those bars, else 0.

Buy the breakout, sell the breakdown — the turtle rule, and the workbench's Donchian bot. Return the list of signals, one per bar from t = window onward.

example 1

input

highs=[10.0, 11.0, 12.0, 13.0], lows=[9.0, 10.0, 11.0, 12.0], closes=[9.5, 10.5, 11.5, 12.9], window=2

output

[1, 1]

example 2

input

highs=[10.0, 11.0, 12.0, 13.0], lows=[9.0, 10.0, 11.0, 12.0], closes=[9.5, 10.5, 11.0, 12.5], window=2

output

[0, 1]

note

The third close sits exactly on the prior low — inside the channel, a 0.

constraints

1 ≤ window < len(closes) ≤ 10_000
lows ≤ closes ≤ highs elementwise

teacher · 2 hints · the first is free

↗ read the theory

the chapter this problem is cut from

▶ open in the workbench

run the same math on a real tape

●solution.py· utf-8 · spaces:4

def donchian_signals(highs: list, lows: list, closes: list, window: int) -> list:
    """1 breakout, -1 breakdown, 0 inside the channel — from t = window onward."""
    out = []
    for t in range(window, len(closes)):
        hi = max(highs[t - window:t])
        lo = min(lows[t - window:t])
        # TODO: touching the channel is not a breakout — the comparisons must be strict
        out.append(1 if closes[t] >= hi else -1 if closes[t] <= lo else 0)
    return out

ln 1, col 1tab indent · esc leave editor

python 3.13 · pyodide

+ 6 hidden on submit

input · keyword arguments to donchian_signals()

highs=[10.0, 11.0, 12.0, 13.0], lows=[9.0, 10.0, 11.0, 12.0], closes=[9.5, 10.5, 11.5, 12.9], window=2

expected

[1, 1]

the judge calls donchian_signals() with exactly these inputs · outputs are compared to 6 decimal places

● standing bypython 3.13 in your browser · pyodide 0.28.3 · judged on the server to 6 dp⌘↵ run · ⌘⇧↵ submit

paper only · educational · not advice