Probabilities.GammaCDFInv Method

Overload List

#SignatureDescription
1procedure GammaCDFInv(const P: TDenseMtxVec; A: Double; B: Double; const Res: TDenseMtxVec);Gamma distribution PPF (vectorized).
2function GammaCDFInv(p: Double; a: Double; b: Double): Double;Gamma distribution inverse CDF (quantile / point percent function, PPF).

Overload 1: procedure GammaCDFInv(const P: TDenseMtxVec; A: Double; B: Double; const Res: TDenseMtxVec);

Gamma distribution PPF (vectorized).

#NameDescription
1PDefines distribution probabilities, real vector or matrix with values within closed interval [0,1].
2ADefines distribution scale parameter. A must be a positive scalar.
3BDefines distribution shape parameter. B must be a positive scalar.
4ResAfter calculation stores the PPF calculated from P, A and B. Length and Complex properties of Res are adjusted automatically to match Length and Complex properties of P.

Result: stored in self (calling object)

Overload 2: function GammaCDFInv(p: Double; a: Double; b: Double): Double;

Gamma distribution inverse CDF (quantile / point percent function, PPF).

#NameDescription
1pProbability, real value on the closed interval [0,1].
2aShape parameter, real value > 0.
3bScale parameter, real value > 0.

Returns: Double - the value x with GammaCDF(x,a,b)=p for SHAPE parameter a and SCALE parameter b. Returns NAN when p < 0, p > 1, a <= 0 or b <= 0; returns 0 at p=0 and 1 at p=1.

Remarks:

Computes the inverse of the Gamma CDF (the quantile function)

inverse CDF(p| a,b ) = F^(-1)(p| a,b) , p(x)=CDF(x|a,b)

evaluated by a Newton-Raphson iteration on P(a,x/b)pP(a,x/b)-p, where Gamma is the Math387.Gamma function. Domain: p in [0,1], a > 0, b > 0. Behaviour: returns 0 at p=0 and 1 at p=1; for invalid parameters or p not in [0,1] returns NAN.

See Also: Probabilities.GammaPDF, Probabilities.GammaCDF