17.04.2013 Views

Distribuţia Binomială: Modelare Statistică, Optimizare Numerică, cu ...

Distribuţia Binomială: Modelare Statistică, Optimizare Numerică, cu ...

Distribuţia Binomială: Modelare Statistică, Optimizare Numerică, cu ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Lorentz JÄNTSCHI (principal investigator) & Sorana D. BOLBOACĂ (co-investigator)<br />

CIWilson_C(X,m) =<br />

2<br />

z 1<br />

X + ±<br />

2 2<br />

2<br />

m + z<br />

± z<br />

38(157)<br />

z<br />

2<br />

−<br />

4<br />

1<br />

m<br />

⎛ X −1<br />

⎞<br />

+ X⎜1−<br />

⎟ ±<br />

⎝ m ⎠<br />

÷ Algoritmi de cal<strong>cu</strong>l:<br />

function Wilson_N(){<br />

tX = this->X + this->z2/2.0; tn = this->n + this->z2;<br />

t0 = this->z * pow(Wald_0+this->z2/4.0,0.5);<br />

this->Xi = (tX-t0)/tn; this->Xs = (tX+t0)/tn;<br />

}<br />

function Wilson_C(){<br />

t0 = 2.0*this->X+this->z2;t1 = 2.0*(this->n+this->z2);<br />

t2 = this->z2-1.0/this->n;<br />

this->Xi = (this->X==0 ? 0 :<br />

(t0-this->z*pow(t2+4*this->X*(1-(this->X-1)/this->n)-2,0.5)-1)/t1);<br />

this->Xs = (this->X==this->n ? 1 :<br />

(t0+this->z*pow(t2+4*this->X*(1-(this->X+1)/this->n)+2,0.5)+1)/t1);<br />

}<br />

÷ Formule matematice:<br />

Intervalul de încredere Agresti-Coull<br />

CIA_C__N(X,m) = CIWald_N(X+z 2 /2,m+z 2 )<br />

CIA_C__C(X,m) = CIWald_N(X+z 2 /4,m+z 2 /2)<br />

CIA_C__D(X,m) = CIWald_N(X+1+4X(m-X)/m,m+4)<br />

÷ Algoritmi de cal<strong>cu</strong>l:<br />

function A_C__N(){<br />

this->Wald_1(this->X+this->z2/2,this->n+this->z2);<br />

}<br />

function A_C__C(){<br />

this->Wald_1(this->X+this->z2/4,this->n+this->z2/2);<br />

}<br />

function A_C__D(){<br />

t = Wald_0/this->n;<br />

Wald_1(this->X+1+4*t,this->n+4);<br />

}<br />

÷ Formule matematice:<br />

Intervalul de încredere ArcSine<br />

CIArcS_N(X,m) = sin 2 (arcsin(√(X/m)±z/√(4n))<br />

CIArcS_C(X,m) = sin 2 (arcsin(√((X+3/8)/(m+3/4))±z/√(4n))<br />

CIArcS_D(X,m) = sin 2 (arcsin(√((X±1/2)/m)±z/√(4n))<br />

CIArcS_E(X,m) = CIArcS_C(X,m) = sin 2 (arcsin(√((X+3/8±1/2)/(m+3/4))±z/√(4n))<br />

÷ Algoritmi de cal<strong>cu</strong>l:<br />

function ArcS_N(){<br />

tX = asin(pow(this->X_n,0.5));<br />

t0 = this->z/pow(4*this->n,0.5);<br />

1<br />

2

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!