rStats.js v1.3.0
    Preparing search index...

    Interface ChisqTestResult

    Result of chisqTest. Has no alternative, because R's chisq.test() does not return one.

    interface ChisqTestResult {
        confInt?: [number, number];
        confLevel?: number;
        estimate?: number | number[];
        estimateNames?: string | string[];
        expected: number[] | number[][];
        ftResiduals: number[] | number[][];
        method: string;
        nullValue?: number;
        nullValueName?: string;
        observed: number[] | number[][];
        parameter: number;
        parameterName?: string;
        pValue: number;
        residuals: number[] | number[][];
        statistic: number;
        statisticName: string;
        stdres: number[] | number[][];
        warnings: string[];
    }

    Hierarchy (View Summary)

    Index
    confInt? confLevel? estimate? estimateNames? expected ftResiduals method nullValue? nullValueName? observed parameter parameterName? pValue residuals statistic statisticName stdres warnings
    confInt?: [number, number]

    Confidence interval [lower, upper] (R's conf.int).

    confLevel?: number

    Confidence level of confInt (R's attr(conf.int, "conf.level")).

    estimate?: number | number[]

    Estimate(s) of the parameter being tested (R's estimate).

    estimateNames?: string | string[]

    Name(s) of estimate (R's names(estimate)), e.g. "mean of x".

    expected: number[] | number[][]

    Expected counts under the null hypothesis.

    ftResiduals: number[] | number[][]

    Freeman-Tukey deviates sqrt(O) + sqrt(O + 1) - sqrt(4E + 1) (Bishop, Fienberg and Holland 1975; not in R), approximately standard normal under the null hypothesis; their sum of squares is the Freeman-Tukey statistic.

    method: string

    Description of the test, as R's method (e.g. "Welch Two Sample t-test").

    nullValue?: number

    Value of the parameter under the null hypothesis (R's null.value).

    nullValueName?: string

    Name of nullValue (R's names(null.value)), e.g. "difference in means".

    observed: number[] | number[][]

    Observed counts (an array of rows for a contingency table).

    parameter: number

    Degrees of freedom (parameterName is "df").

    parameterName?: string

    Name of parameter (e.g. "df").

    pValue: number

    p-value of the test (R's p.value).

    residuals: number[] | number[][]

    Pearson residuals (O - E) / sqrt(E).

    statistic: number

    Value of the test statistic (R's statistic).

    statisticName: string

    Name of the test statistic (e.g. "t", "X-squared", "V", "W").

    stdres: number[] | number[][]

    Standardized residuals (O - E) / sqrt(V), where V is E (1 - row proportion) (1 - column proportion) for a contingency table and n p (1 - p) for goodness of fit. For a contingency table the complements are the other rows' and columns' sums over the total, and the residual and V are computed in double-double.

    warnings: string[]

    R's warnings, each once, in R's order (an empty array when there are none).

    Settings
    Member Visibility

    On This Page
    Properties