function 1 : log-likelihood to estimate logmu
f(logmu)
= function 2 : log-likelihood to estimate beta
f(beta)
=
= data in group1
= data in group2
for
,
for
based on 10^6 time runs
of
,
of
, mean
, variance
of ,
, mean
, variance
of without NAN and negative valuesmean
= / n variance
= 1% quantile
= value at (pos-1) + diff_v*diff_pcoefficients
: residuals
= y - (intercept + slope * x) xd2hat
= variance of residuals - Var xinitial
= estimate value of nlm
lkdinitial
min
and estimate
of xinitiallkdinf
xbeta.mle
= xinitial[(m+1):(m+2)]xbar.mle
= exponent of xinitialxs2.mle
= exponent of xbeta.mle * xbar.mle ^ xbeta.mle