Cumulative Distribution Function for LMS Quantile Regression
Computes the cumulative distribution function (CDF) for observations, based on a LMS quantile regression.
cdf.lmscreg(object, newdata = NULL, ...)
object |
A VGAM quantile regression model, i.e.,
an object produced by modelling functions such as |
newdata |
Data frame where the predictions are to be made. If missing, the original data is used. |
... |
Parameters which are passed into functions such as
|
The CDFs returned here are values lying in [0,1] giving the relative
probabilities associated with the quantiles newdata.
For example, a value near 0.75 means it is close to the upper quartile
of the distribution.
A vector of CDF values lying in [0,1].
The data are treated like quantiles, and the percentiles
are returned. The opposite is performed by
qtplot.lmscreg.
The CDF values of the model have been placed in
@post$cdf when the model was fitted.
Thomas W. Yee
Yee, T. W. (2004). Quantile regression via vector generalized additive models. Statistics in Medicine, 23, 2295–2315.
fit <- vgam(BMI ~ s(age, df=c(4, 2)), lms.bcn(zero = 1), data = bmi.nz) head(fit@post$cdf) head(cdf(fit)) # Same head(depvar(fit)) head(fitted(fit)) cdf(fit, data.frame(age = c(31.5, 39), BMI = c(28.4, 24)))
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.