OpenBB-finance/OpenBB · error · ValueError

Error: lower_q and upper_q must be between 0 and 1

Error message

Error: lower_q and upper_q must be between 0 and 1

What it means

In the cones helper (openbb_technical/helpers.py), after normalizing lower_q/upper_q so lower <= upper, a hard check rejects quantile values >= 1. Quantiles must be strictly within [0, 1) as fractions, not percentages.

Source

Thrown at openbb_platform/extensions/technical/openbb_technical/helpers.py:419

        "parkinson",
        "garman_klass",
        "hodges_tompkins",
        "rogers_satchell",
        "yang_zhang",
    ],
    trading_periods: int | None = None,
) -> "DataFrame":
    """Calculate Cones."""
    # pylint: disable=import-outside-toplevel
    from pandas import DataFrame

    estimator = DataFrame()

    if lower_q > upper_q:
        lower_q, upper_q = upper_q, lower_q

    if (lower_q >= 1) or (upper_q >= 1):
        raise ValueError("Error: lower_q and upper_q must be between 0 and 1")

    lower_q_label = str(int(lower_q * 100))
    upper_q_label = str(int(upper_q * 100))
    quantiles = [lower_q, upper_q]
    windows = [3, 10, 30, 60, 90, 120, 150, 180, 210, 240, 300, 360]
    min_ = []
    max_ = []
    median = []
    top_q = []
    bottom_q = []
    realized = []
    allowed_windows = []
    data = data.sort_index(ascending=True)

    model_functions = {
        "std": standard_deviation,
        "parkinson": parkinson,
        "garman_klass": garman_klass,

View on GitHub (pinned to 3e071fcc2c)

Solutions

1. Express quantiles as fractions: lower_q=0.10, upper_q=0.90.
2. Divide percentage-style values by 100 before the call.
3. Note bounds are auto-swapped if reversed, so only the 0-1 range matters.
4. Validate inputs at the config layer of your app so users cannot enter percent integers.

Example fix

# before
cones(data, lower_q=10, upper_q=90)  # interpreted as >= 1

# after
cones(data, lower_q=0.10, upper_q=0.90)

assert 0 <= lower_q < 1 and 0 <= upper_q < 1, "quantiles must be fractions in [0, 1)"
lower_q, upper_q = sorted([lower_q, upper_q])

def valid_quantile(q) -> bool:
    return isinstance(q, (int, float)) and 0 <= q < 1

try:
    out = cones(data, lower_q=lower_q, upper_q=upper_q)
except ValueError as e:
    if "must be between 0 and 1" in str(e):
        out = cones(data, lower_q=lower_q/100, upper_q=upper_q/100)
    else:
        raise

When it happens

Trigger: Calling cones(data, lower_q=0.25, upper_q=95) style code where one bound is passed as a percentage (>= 1) instead of a fraction; also lower_q=1 or upper_q=1 exactly.

Common situations: Translating percentile integers (e.g. '95th percentile') directly into the parameter; copying configs from libraries that accept 0-100 quantiles; mixing conventions between lower_q and upper_q.

Related errors

• Data length is less than required by parameters: {max(length
• Column {close_col} not in data
• At least one extension type must be selected.
• Incorrect email or password
• Charting is not installed. Please install `openbb-charting`.

AI-assisted analysis of OpenBB-finance/OpenBB@3e071fcc2c (2026-08-14). Data as JSON: /api/errors/4d539c5b6d1b63ac. Report an issue: GitHub.