MAIDR Documentation - v4.4.0
    Preparing search index...

    Interface SurvivalPoint

    One point of a Kaplan-Meier survival curve.

    The curve itself is a step function -- survival holds until an event drops it -- so this is a StepPoint with the two things a survival figure carries that a step chart does not.

    interface SurvivalPoint {
        x: string | number;
        y: number | null;
        z?: string;
        label?: string;
        censored?: boolean;
        yMin?: number;
        yMax?: number;
    }

    Hierarchy (View Summary)

    Index

    Properties

    x: string | number
    y: number | null

    The magnitude at this x, or null where the series has a position but no reading.

    A gap is not a zero. seaborn.pointplot pads a hue level missing from one category so its estimate lines stay the same length, and a producer that meets a break in a series has nowhere else to put it. Number(null) is 0, which would make the gap sound like a real low point, let it be reached as the row's minimum, and pull the range every other point's pitch is scaled against — the same trap toBarValue was written for (#925).

    null rather than a non-finite number because the payload has to survive JSON.parse: json.dumps writes NaN and Infinity as bare tokens that are legal JavaScript and invalid JSON, and a producer emitting one stops the chart initialising at all (xability/py-maidr#427).

    z?: string
    label?: string

    Ordinal level name announced in place of the raw numeric y, for a chart whose y axis is a category rather than a magnitude — a hypnogram's sleep stages, a Likert response, a severity grade. y stays numeric because it drives sonification, braille and the min/max range, so the human-readable name has to travel alongside it.

    An empty string counts as absent, so a producer that emits '' for an unnamed level gets the numeric announcement rather than a blank one. Omitting it entirely is the right shape for a continuous y.

    { x: 1.5, y: 3, label: 'REM' }
    
    censored?: boolean

    A subject left the study at this time without the event happening.

    Censoring marks are drawn as ticks on the curve rather than as steps, because censoring does not change the estimate -- it changes how much of the curve is still supported by data. A reader who cannot tell a censored time from an ordinary one cannot tell a flat tail backed by two hundred subjects from one backed by three.

    yMin?: number

    Lower bound of the confidence band at this time, when the chart draws one.

    yMax?: number

    Upper bound of the confidence band at this time, when the chart draws one.