ts-math/lib/Math.d.ts

A collection of math functions and packages written in Typescript

export declare const sqr: (x: number) => number;
export declare function sum(xs: number[]): number;
export declare function mean(xs: number[]): number;
/**
 * Calculates the standard deviation of samples in the input array
 * ```
 * \operatorname{stdev}(x)=\sqrt{\frac{\sum{x^2}-\frac{\sum{x}}{n}}{n-1}}
 * ```
 */
export declare function stdev(xs: number[]): number;
/**
 * The corr function returns the correlation coefficient of two
 * ```
 * \operatorname{corr}(x,y)=\frac{n\sum{xy}-\sum{x}\sum{y}}{\sqrt{(n\sum{x^2}-(\sum{x})^2)(n\sum{y^2}-(\sum{y})^2)}}
 * ```
 */
export declare function corr(xs: number[], ys: number[]): number;
/**
 * The cov function calculates the covariance between the two arrays.
 * ```
 * \operatorname{cov}(x,y)=\frac{\sum{xy}-\frac{\sum{x}\sum{y}}{n}}{n-1}
 * ```
 */
export declare function cov(xs: number[], ys: number[]): number;
/**
 * In probability theory and statistics, skewness is a measure of the asymmetry of the probability distribution of a real-valued random
 * variable about its mean.
 */
export declare function skew(xs: number[]): number;
/**
 * In probability theory and statistics, kurtosis (from Greek: κυρτός, kyrtos or kurtos, meaning "curved, arching") is a measure of the
 * "tailedness" of the probability distribution of a real-valued random variable.
 */
export declare function kurtosis(xs: number[]): number;
/**
 * erf is the "error function" encountered in integrating the normal distribution (which is a normalized form of the Gaussian function).
 * It is an entire function defined by
 * ```
 * \operatorname{erf}(z) = \frac{2}{\sqrt{\pi}}\int_{0}^z e^{-t^2}dt
 * ```
 */
export declare function erf(z: number): number;
/**
 * The complementary error function
 */
export declare function erfc(x: number): number;
/**
 * Inverse complementary error function
 */
export declare function erfcinv(p: number): number;
/**
 * Normal inverse cumulative distribution function
 * @param {Number} p Probability
 * @param {Number} mu Mean
 * @param {Number} sigma Standard deviation
 */
export declare function normalInv(p: number, mu: number, sigma: number): number;
/**
 * Normal cumulative distribution function
 * @param {*} x Value
 * @param {*} mu Mean
 * @param {*} sigma Standard deviation
 */
export declare function normalCdf(x: number, mu: number, sigma: number): number;
/**
 * Function that returns the q-th quantile of a dataset (same as numpy.quantile in Python)
 */
export declare function quantile(xs: number[], q: number): number;
/**
 * Rounds with decimals. It should round away from zero, therefore it rounds absolute amount. It also adds a small amount to avoid
 * midpoint rounding issues, like 0.4999999999 should round to 1
 */
export declare function round(x: number, decimals: number): number;
/**
 * Generates an array of numbers. Values are set to 0 up to n-1, incrementing by 1.
 * @param n The numbers of array items.
 * @returns array of numbers
 */
export declare function range(n: number): number[];
/**
 * Returns the index of the highest element that is lower or equal to t. The input array must be ordered in ascending order.
 * @param t
 * @param vs
 * @returns
 */
export declare function indexOf(t: number, vs: number[]): number;
/**
 * Linear regression of xs-values against ys-values. Weighted by ws-values (if included).
 * @param xs
 * @param ys
 * @param ws
 * @returns
 */
export declare function linearRegression(xs: number[], ys: number[], ws?: number[] | null): {
    k: number;
    b: number;
    error: number;
};
/**
 * Assumes the vs-values are normally distributed, adds them to a QQ-plot and estimates the mean, stdev by linear regression in the plot.
 * The values in the QQ-plot can be centrally weighted (useful if data can contain anomalies or non-normally extreme jumps).
 * @param vs
 * @param centrallyWeighted true or false
 * @returns
 */
export declare function qqRegression(vs: number[], centrallyWeighted?: boolean): {
    stdev: number;
    mean: number;
    error: number;
};
/**
 * Polynomial regression. Estimates the coefficients of the polynomial expression the fits the xs and ys-values. Weighted by ws-values (if included).
 * @param xs x-values
 * @param ys y-values
 * @param ws weights
 * @returns coefficients array
 */
export declare function polynomialRegression(xs: number[], ys: number[], ws: number[] | null, n: number): number[];