Cperiodogram: Contingency periodogram

In adespatial: Multivariate Multiscale Spatial Analysis

Description

Function to compute a contingency periodogram for a univariate series of qualitative data

Usage

Cperiodogram(x, T1 = 2, T2 = NULL, nperm = NULL, alpha = 0.05,
  graph = TRUE)

Arguments

x	a qualitative variable (factor)
T1	first period included in the calculations (default: T1 = 2)
T2	last period included in the calculations (default: T2 = n/2)
nperm	Number of permutations for the chi-square test. For chi-square tests using the chi- square distribution, use the default nperm=NULL
alpha	significance level for computation of the confidence limits
graph	a logical indicating if a graph is requested, by default TRUE.

Details

The contingency periodogram of Legendre et al. (1981) identifies periodic components in qualitative data vectors. The vector may contain classes of a qualitative variable or the classes obtained by hierarchical clustering or partitioning of a multivariate data table. The method is also described in Legendre & Legendre (2012). The optional graph produced by the function shows the following information:

• In red: the B statistics (information in common).

• In blue: Confidence limits for B without correction.

• In green: Bonferroni-corrected limits of the confidence intervals.

• In black: Confidence limits with progressive Bonferroni correction.

Value

A table with the statistics for the selected periods:

• Wilks<e2><80><99> chi-square statistic (Wilks.chisq)

• information in common (B),

• degrees of freedom (df),

• p-value (prob)

Confidence interval limits:

• critical value of B without correction (B.crit),

• critical value of B with Bonferroni correction based on the number of periods studied in the periodogram (B.crit.Bonf),

• critical value of B with progressive Bonferroni correction (B.prog.Bonf).

Author(s)

Pierre Legendre [e-mail]

References

Legendre, L., M. Fr<c3><a9>chette & P. Legendre. 1981. The contingency periodogram: a method of identifying rhythms in series on nonmetric ecological data. Journal of Ecology 69: 965-979.

Legendre, P. and Legendre, L. 2012. Numerical Ecology. 3rd English ed. Elsevier, Amsterdam

Examples

# Data from the numerical example of Subsection 12.4.2 of Legendre and Legendre (2012).
test.vec <- c(1,1,2,3,3,2,1,2,3,2,1,1,2,3,3,1)
# Periodogram with tests using the chi-square distribution
res <- Cperiodogram(test.vec)
# Periodogram with permutation tests
res <- Cperiodogram(test.vec, nperm=2000, graph=FALSE)

Example output

Registered S3 methods overwritten by 'adegraphics':
  method         from
  biplot.dudi    ade4
  kplot.foucart  ade4
  kplot.mcoa     ade4
  kplot.mfa      ade4
  kplot.pta      ade4
  kplot.sepan    ade4
  kplot.statis   ade4
  scatter.coa    ade4
  scatter.dudi   ade4
  scatter.nipals ade4
  scatter.pco    ade4
  score.acm      ade4
  score.mix      ade4
  score.pca      ade4
  screeplot.dudi ade4
Registered S3 method overwritten by 'spdep':
  method   from
  plot.mst ape 
Registered S3 methods overwritten by 'adespatial':
  method             from       
  plot.multispati    adegraphics
  print.multispati   ade4       
  summary.multispati ade4       
Warning messages:
1: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect
2: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect
3: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect
4: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect
5: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect
6: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect
7: In chisq.test(tab, simulate.p.value = simulate, B = nperm) :
  Chi-squared approximation may be incorrect

adespatial documentation built on May 2, 2019, 4:43 p.m.