Description Data classes, visualisation of data Manipulation and summary functions Tests Author(s) References Examples
The package fdnonpar is a collection of tools for visualisation and nonparametric analysis of grouped data that consist of an independent (x-) variable and a dependent (y-) variable. Such data could be time series or values of a function, or generally any multivariate data. In the latter case, x is just an index variable.
The proposed tests do not take the x-variable into account. Since the x-values are usually equidistant, this does not really matter. In the current version (0.3), it is assumed that the data samples to be compared share the same (equidistant) x-variables.
fdsample | basic data type, a sample of curves |
fdenvelope | two boundary curves of an envelope, a specialized
fdsample object
|
print.fdsample | print brief details of an fdsample |
plot.fdsample | plot the individual members of an fdsample |
plot.fdenvelope | plot an envelope object
|
summaryplot | plot individual curves, envelopes or summary functions |
such as the mean of an fdsample objectr
|
[.fdsample | extract or replace curves form a sample |
apply.fdsample | apply a summary function to the function values |
mean.fdsample | mean of the function values |
median.fdsample | median of the function values |
quantile.fdsample | quantiles of the function values |
pwEnvelope | pointwise envelope |
tL2.permtest | Comparison of two groups, |
using square integrated Welch-t-statistic | |
rankEnv.test | Rank envelope test: compare a single observation to a group (e.g. simulated data) |
rankCount.test | Rank count test: refined p-value for rank envelope test |
Ute Hahn, ute@imf.au.dk
Hahn, U. (2012) A studentized permutation test for the comparison of spatial point patterns. Journal of the American Statistical Association 107 (498), 754–764.
M. Myllymaki, T. Mrkvicka, P. Grabarnik, H. Seijo and Ute Hahn (2015) Global envelope tests for spatial processes, http://arxiv.org/abs/1307.0239v3.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | # simulated data sets, consisting of 8 and 9
x <- seq(0, 1, .1)
y1 <- replicate(8, rnorm(length(x), mean = x, sd = .2))
y2 <- replicate(7, rnorm(length(x), mean = x*1.3, sd = .2))
y1b <- replicate(7, rnorm(length(x), mean = x, sd = .2))
xy1 <- fdsample(x, y1)
xy2 <- fdsample(x, y2)
xy1b <- fdsample(x, y1b)
# visualize the data sets
# require plutils which contains the generic to "summaryplot"
require(plutils)
summaryplot(xy1, envprob=1)
summaryplot(xy1b, add = TRUE, col = "blue")
summaryplot(xy2, add = TRUE, col = "red")
# there should be significant difference between xyl1
# and xyl2, but not between xyl1 and xyl1b.
# However, with simulated data, everything is possible...
tL2.permtest(xy1, xy2)
tL2.permtest(xy1, xy1b)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.