G1G2G3(link = "logit", ip1 = NULL, ip2 = NULL, iF = NULL)
p1
, p2
and f
.
See Links
for more choices.p1
, p2
and f
."vglmff"
(see vglmff-class
).
The object is used by modelling functions such as vglm
and vgam
.p1
and p2
are probabilities, so that
p3=1-p1-p2
is the third probability.
The parameter f
is the third independent parameter.AA.Aa.aa
,
AB.Ab.aB.ab
,
AB.Ab.aB.ab2
,
ABO
,
MNSs
.y <- cbind(108, 196, 429, 143, 513, 559)
fit <- vglm(y ~ 1, G1G2G3(link = probit), trace = TRUE, crit = "coef")
fit <- vglm(y ~ 1, G1G2G3(link = logit, ip1 = 0.3, ip2 = 0.3, iF = 0.02),
trace = TRUE, crit = "coef")
fit <- vglm(y ~ 1, G1G2G3(link = "identity"), trace = TRUE)
Coef(fit) # Estimated p1, p2 and f
rbind(y, sum(y)*fitted(fit))
sqrt(diag(vcov(fit)))
Run the code above in your browser using DataLab