Overload List
| # | Signature | Description |
|---|---|---|
| 1 | procedure FatigueLifeCDFInv(const P: TDenseMtxVec; Mu: Double; gamma: Double; beta: Double; const Res: TDenseMtxVec); | Fatigue Life PPF (vectorized). |
| 2 | function FatigueLifeCDFInv(p: Double; Mu: Double; gamma: Double; beta: Double): Double; | Fatigue life (Birnbaum-Saunders) percent point function (PPF, quantile / inverse CDF). |
Overload 1: procedure FatigueLifeCDFInv(const P: TDenseMtxVec; Mu: Double; gamma: Double; beta: Double; const Res: TDenseMtxVec);
Fatigue Life PPF (vectorized).
| # | Name | Description |
|---|---|---|
| 1 | P | Defines distribution probabilities, real vector or matrix with values within closed interval [0,1]. |
| 2 | Mu | Defines distribution location parameter. |
| 3 | gamma | Defines distribution shape parameter. gamma must be a positive scalar. |
| 4 | beta | Defines distribution scale parameter. beta must be a positive scalar. |
| 5 | Res | After calculation stores the PPF calculated from P, m, gamma and beta. 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 FatigueLifeCDFInv(p: Double; Mu: Double; gamma: Double; beta: Double): Double;
Fatigue life (Birnbaum-Saunders) percent point function (PPF, quantile / inverse CDF).
| # | Name | Description |
|---|---|---|
| 1 | p | Probability, real value on the closed interval [0,1]. |
| 2 | Mu | Distribution location parameter, real value. |
| 3 | gamma | Distribution shape parameter. gamma must be a positive scalar (gamma>0). |
| 4 | beta | Distribution scale parameter. beta must be a positive scalar (beta>0). |
Returns: Double - the value x such that FatigueLifeCDF(x,Mu,gamma,beta)=p. Parameters gamma and beta must be positive and the probability p must lie on [0,1], otherwise the result is NaN.
Remarks:
The inverse CDF has a closed form. Let w=gamma Phi^(-1)(p) where Phi^(-1) is the standard normal quantile; then with z=(x-mu)/beta the relation w=sqrt(z)-sqrt(1/z) inverts to a quadratic in z:
CDFInv(p | mu,gamma,beta) = mu + beta z , z = 1/2( w^2+2 +/- sqrt((w^2+2)^2-4) ) , w=gamma Phi^(-1)(p) .
The "+" root is taken for and the "-" root for . Returns NaN for p outside [0,1], gamma<=0 or beta<=0.