columna API reference - v0.5.0
    Preparing search index...

    Interface GlmResult

    interface GlmResult {
        aic: number;
        aicc: number;
        bic: number;
        coefficients: GlmCoefficient[];
        confidence: number;
        converged: boolean;
        cooksD: Float64Array;
        covariance: Matrix;
        deviance: number;
        dfNull: number;
        dfResidual: number;
        dispersion: number;
        family: Family;
        fitted: Float64Array;
        goodnessOfFit: {
            deviance: { df: number; pValue: number; statistic: number };
            hosmerLemeshow?: {
                df: number;
                groups: { expected: number; n: number; observed: number }[];
                pValue: number;
                statistic: number;
            };
            pearson: { df: number; pValue: number; statistic: number };
        };
        gTest: { df: number; pValue: number; statistic: number };
        iterations: number;
        leverage: Float64Array;
        linearPredictor: Float64Array;
        link: Link;
        logLik: number;
        n: number;
        names: string[];
        nullDeviance: number;
        omitted: number[];
        p: number;
        pearsonChi2: number;
        r2deviance: number;
        r2devianceAdj: number;
        residuals: {
            deviance: Float64Array;
            pearson: Float64Array;
            response: Float64Array;
            standardizedPearson: Float64Array;
        };
        test: "glm";
        predict(
            x: ArrayLike<number> | ArrayLike<number>[],
            options?: { confidence?: number; type?: "link" | "response" },
        ): { ci: [number, number]; fit: number; se: number }[];
    }
    Index
    aic aicc bic coefficients confidence converged cooksD covariance deviance dfNull dfResidual dispersion family fitted goodnessOfFit gTest iterations leverage linearPredictor link logLik n names nullDeviance omitted p pearsonChi2 r2deviance r2devianceAdj residuals test
    predict
    aic: number
    aicc: number
    bic: number
    coefficients: GlmCoefficient[]
    confidence: number
    converged: boolean
    cooksD: Float64Array
    covariance: Matrix
    deviance: number
    dfNull: number
    dfResidual: number
    dispersion: number

    Estimated dispersion (1 for binomial / poisson).

    family: Family
    fitted: Float64Array
    goodnessOfFit: {
        deviance: { df: number; pValue: number; statistic: number };
        hosmerLemeshow?: {
            df: number;
            groups: { expected: number; n: number; observed: number }[];
            pValue: number;
            statistic: number;
        };
        pearson: { df: number; pValue: number; statistic: number };
    }

    Goodness-of-fit tests (deviance, Pearson; Hosmer–Lemeshow for binary logistic).

    gTest: { df: number; pValue: number; statistic: number }

    Deviance-based test that all slopes are zero (Minitab "G").

    iterations: number
    leverage: Float64Array
    linearPredictor: Float64Array
    link: Link
    logLik: number
    n: number
    names: string[]
    nullDeviance: number
    omitted: number[]
    p: number
    pearsonChi2: number
    r2deviance: number

    Deviance R² and adjusted (Minitab: 1 − dev/nullDev).

    r2devianceAdj: number
    residuals: {
        deviance: Float64Array;
        pearson: Float64Array;
        response: Float64Array;
        standardizedPearson: Float64Array;
    }
    test: "glm"

    • Parameters

      • x: ArrayLike<number> | ArrayLike<number>[]
      • Optionaloptions: { confidence?: number; type?: "link" | "response" }

      Returns { ci: [number, number]; fit: number; se: number }[]

    Settings
    Member Visibility

    On This Page
    Properties
    Methods