cutoff | R Documentation |
cutoff
returns the cutoff value from a bimodal distribution, together
with its confidence interval.
cutoff(object, t = 1e-64, nb = 10, distr = 2, type1 = 0.05, level = 0.95)
object |
An output from the function |
t |
A numerical scalar indicating the value below which the E-M algorithm should stop. |
nb |
Number of Monte Carlo simulations. |
distr |
Either 1 or 2, indicates which distribution belonging the Type-I
error corresponds to. 1 correspond to the first distribution in |
type1 |
A numerical value between 0 and 1, the value of the type-I error. |
level |
The confidence level required. |
From a fitted finite mixture model, we compute the probability to belong to
one of the two probability distributions of the finite mixture model, as a
function of the datum value. This probability function is used to look for the
cutoff value defined as the datum value for which the probability to belong to
a given probability distribution is equal to a type-I error. The confidence
interval of this cutoff value is computed by Monte Carlo simulations where in
each iteration the five parameter values of the finite mixture model are
sampled in a multinormal distribution and then the cutoff value is computed.
The confidence interval of the cutoff value is computed by fitting a normal
distribution by maximum likelihood to the Monte-Carlo-derived values of the
cutoff. This last step is performed by the fitdistr
function of the
MASS
package.
Returns a numerical vector of lenght 3, the first value being the
estimated value of the cutoff and the second and thrid values being the
lower and upper bound of the confidence interval of this estimate, at the
level specified by parameter level
.
Trang N.V., Choisy M., Nakagomi N.T., Chinh N.T.M., Doan, Y.H., Yamashiro T., Bryant J.E., Nakagomi O. and Anh D.D. (2015) Determination of cut-off cycle threshold values in routine RT-PCR assays to assist differential diagnosis of norovirus in children hospitalized for acute gastroenteritis. Epidemiol. Infect. In press.
em
for fitting a finite mixture model.
# Measles IgG concentration data: length(measles) range(measles) # Plotting the data: hist(measles,100,F,xlab="concentration",ylab="density",ylim=c(0,.55), main=NULL,col="grey") # Estimating the parameters of the finite mixture model: (measles_out <- em(measles,"normal","normal")) # Adding the E-M estimated finite mixture model: lines(measles_out,lwd=1.5,col="red") # Estimating a cutoff value from this fitted finite mixture model: (cut_off <- cutoff(measles_out)) polygon(c(cut_off[-1],rev(cut_off[-1])),c(0,0,.55,.55), col=rgb(0,0,1,.2),border=NA) abline(v=cut_off[-1],lty=2,col="blue") abline(v=cut_off[1],col="blue")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.