| pulini {unknown} | R Documentation |
This function accomplishes the first step of the algorithm in Yang, Liu and Wang (2004). It finds potential pulse locations. The function itself can also be used for pulse detection.
pulini(x, y, data, method = c("pcp", "CLUSTER"), alpha,
control=list(pcp=list(spline=list(nb=~x, rk=cubic(x)), spar="v",
limnla=c(-10, 3)), cluster=list(sd=mean(y)*0.07, nnadir=2, npeak=3)))
x |
a vector of observation time points. |
y |
a vector of hormone concentrations. |
data |
a data frame containing the variables occurring in the x and y arguments. If this option is not specified, the variables should be on the search list. Missing values are not allowed. |
method |
the method to be used for identifying initial pulse locations. If “pcp”, the change point method based on partial smoothing spline models is used to detect pulse locations as change points to the first derivative of the mean function. If “CLUSTER”, the CLUSTER method proposed by Veldhuis and Johnson (1986) is used. |
alpha |
for method="pcp", alpha controls the significance level of a potential change point; for method="CLUSTER", alpha controls the significance level of the $t$ test. |
control |
A list of two components, pcp and cluster,
to replace the default values in the pcp and
CLUSTER functions. |
pulini is a wrapper of two other functions,
pcp and
CLUSTER. See these two
functions for details about control otions. Larger
alpha leads to more identified pulses, thus increases
false positive rate and decreases false negative rate.
CLUSTER is faster than pcp,
however, its false negative rate is usually a bit larger.
a vector of pulse locations.
Yu-Chieh Yang, Anna Liu, Yuedong Wang
Veldhuis, J. D. and Johnson, M. L. (1986), Cluster analysis: a simple versatile and robust algorithm for endocrine pulse detection, American Journal of Physiology, 250, E486-E493.
Yang, Y. (2002), Detecting Change Points and Hormone Pulses Using Partial Spline Models, Ph.D. Thesis, University of California-Santa Barbara, Dept. of Statistics and Applied Probability.
Yang, Y. and Liu, A. and Wang, Y., (2004), Detecting Pulsatile Hormone Secretions Using Nonlinear Mixed Effects Partial Spline Models. Available at www.pstat.ucsb.edu/faculty/yuedong/research.
pl1 <- pulini(time, conc, data=acth, method="pcp", alpha=0.6)
pl2 <- pulini(time, conc, data=acth, method="CLUSTER", alpha=.2,
control=list(cluster=list(sd=.05*mean(acth$conc))))