Estimating the (single) parameter of the 1-truncated logarithmic distribution.
otlog(lshape = "logitlink", gshape = ppoints(8), zero = NULL)
Same as
logff
.
An object of class "vglmff"
(see vglmff-class
).
The object is used by modelling functions such as vglm
,
and vgam
.
The 1-truncated logarithmic distribution is a logarithmic distribution but with
the probability of a one being zero. The other probabilities are scaled
to add to unity.
Some more details can be found at logff
.
Multiple responses are permitted.
# NOT RUN {
odata <- data.frame(y1 = rotlog(n = 1000, shape = logitlink(1/3, inverse = TRUE)))
ofit <- vglm(y1 ~ 1, otlog, data = odata, trace = TRUE, crit = "c")
coef(ofit, matrix = TRUE)
Coef(ofit)
# }
# NOT RUN {
with(odata,
hist(y1, shape = TRUE, breaks = seq(0.5, max(y1) + 0.5, by = 1),
border = "blue"))
x <- seq(1, with(odata, max(y1)), by = 1)
with(odata, lines(x, dotlog(x, Coef(ofit)[1]), col = "orange", type = "h", lwd = 2))
# }
Run the code above in your browser using DataLab