4af6d330-66a2-4b46-ac51-064b578f95b7 60 2 5 false false 0 decimal

&[DATE] &[TIME] - &[FILENAME]

If this is slightly increased to 13.25, we get α > 1, and no error

]]>

X 30.4 13.22 63.9 387.1 330 187.3 495.9 18 163.3 128.8 101 mat :

3 appVersion 0.99.7251

278.7

13.22

X.mean X Mean : 181.792

LNX X ln vectorize :

LNX.mean LNX Mean : 4.619457599

A X.mean ln LNX.mean - : 0.583405577

α 14 A * / 114 A * 3 / + sqrt + (* eval : 0.9999

Works correct only if numerically evaluated.

Otherwise gives different results in evaluating F & G when using the vectorize operater ( only if α ≥ 1 )

]]>

β X.mean α / eval : 181.8117

α 0.9999

If 0 < α < 1 in Eqn 1 throwss an error . This behaviour seems to be a bug, either

in SS 7251 OR evaluation of Eqn 1. But it evaluates correctly in EXCEL

]]>

z F 1 β α^α Gamma * / x α 1 -^e x β / -^* x 0 z int * :

Eqn 1

]]>

Program 1 j 1 X rows range G j el X j el F : eval 11 line for

X 3 el F #

lastError x - not defined.

G #

X F vectorize #

As per Jean's method

]]>

ε 1015 -^:

This solves the problem

]]>

eval(,) drops timing ~ 0.75
acc acuuracy 50/100 plentty. x α β CDF 1 α Gamma β α^* / t α 1 -^t β / - exp * t ε x int eval * :

Mathcad result

]]>

Excel Result

]]>

X α β CDF vectorize 0.154 0.0702 0.2964 0.8811 0.8372 0.6431 0.9346 0.0943 0.5927 0.5076 101 mat