netCorr produces a network object of dynamic correlations. Its input has to be at least one set of quantitative variables.
d_netCorr(variables, nodes = NULL, weight=NULL,
pairwise=FALSE, minimum=-Inf, maximum=Inf,
frequency=FALSE, means=TRUE,
method=c("pearson", "kendall", "spearman"), criteria="value", Bonferroni=FALSE,
minL=0, maxL=Inf,
sortL=NULL, decreasingL=TRUE,
factorial=c("null", "pc", "nf", "vf", "of"),
components=TRUE, backcomponents=FALSE,
sequence=seq(.20, 1, .01), textFilter=c(1, .99), speed=50,
dir=NULL, ...)The function creates a list object to create a multigraph and eventually a folder in the computer with an HTML document named index.html which contains the produced dynamic graphs. This file can be directly opened with your browser and sent to a web server to work properly.
a data frame with at least two quantitative variables.
a data frame with at least one vector of names and other information from the nodes
a vector of weights. Optimal for data.framed tables
Pairwise mode of handling missing values if TRUE. Listwise by default.
minimum mean to be considered
maximum mean to be considered
a logical value true if frequencies are to be shown. Default=FALSE.
a logical value true if means are to be shown. Default=TRUE.
a vector of statistics of similarity. Pearson correlation by default. Spearman and Kendall are also possible
statistic to be use for selection criteria.
Bonferroni criterium of the signification test.
minimum value of the statistic to include the edge in the list.
maximum value of the statistic to include the edge in the list.
sort the list according to the values of a statistic. See below
order in a decreasing way.
factorial layout: Principal components (pc), factorial (nf), factorial with varimax rotation (vf), and factorial with oblimin rotation (of) are possible
display following graph in sequence only when the components are unequal
display following graph in sequence even if there are less components
evolution of the dinamic graphs. c(first threshold, last threshold, step)
limits for showing the correlations as text on the graph. This limit is the minimun value of the first (absolute), and the second (threshold plus its value)
speed of the dynamic evolution from 0 to 100
a "character" string representing the directory where the web files will be saved.
any netCoin argument.
Modesto Escobar, Department of Sociology and Communication, University of Salamanca. See https://sociocav.usal.es/blog/modesto-escobar/
# A character column (with separator)
data(iris)
d_netCorr(iris[,1:4],ltext="value", sequence=seq(.80, 1, .05), zoom=3, textFilter=.85,
main="Correlations between measurements of Iris Species", components=FALSE,
note="Anderson, Edgar (1935) y Fisher, R. A. (1936)") # network object
Run the code above in your browser using DataLab