A function to classify continuous variables.
This function is a wrapper for
classIntervals
with some additional methods.
mf_get_breaks(x, nbreaks, breaks, k = 1, central = FALSE, ...)A numeric vector of breaks
a vector of numeric values. NA and Inf values are not used in the classification.
a number of classes
a classification method; the main methods are "quantile", "equal", "msd", "ckmeans" (natural breaks), "Q6" and "geom". See Details for the full list.
number of standard deviation for "msd" method (see Details)
creation of a central class for "msd" method (see Details)
further arguments
of classIntervals
classInt methods
"fixed", "sd", "equal", "pretty", "quantile", "kmeans", "hclust", "bclust",
"fisher", "jenks", "dpih", "headtails", "maximum", and "box"
are classIntervals
methods. You may need to pass additional arguments for some of them.
The "jenks", "fisher" and "ckmeans" methods are based on the same concept of natural breaks and and produce similar groupings. The use of "ckmeans" is recommended.
The "jenks" method produces class boundaries falling on data points and is slow.
The "fisher" method produces class boundaries located more conveniently between data points, and is faster than the "jenks" method.
The "ckmeans" method produces exactly the same class boundaries as the
"fisher" method, but is much faster. It uses the optimal univariate
k-means method from the Ckmeans.1d.dp package.
If the "ckmeans" method is selected but the Ckmeans.1d.dp package is not
installed then the "fisher" method is used.
The relative speeds of these three methods may vary depending on the number of data points and the number of classes.
The "msd" method is based on the mean and the standard deviation
of a numeric vector.
The nbreaks parameter is not relevant, use k and
central instead. k indicates
the extent of each class in share of standard deviation.
If central=TRUE then
the mean value is the center of a class else the mean is a break value.
The "q6" method uses the following quantile
probabilities: 0, 0.05, 0.275, 0.5, 0.725, 0.95, 1.
The "Q6" method uses the following quantile
probabilities: 0, 0.05, 0.25, 0.5, 0.75, 0.95, 1.
The "geom" method is based on a geometric progression along
the variable values, all values must be strictly greater than zero.
The "arith" method is based on an arithmetic progression along
the variable values.
The "em" method is based on nested averages computation.
Breaks defined by a numeric vector or a classification method are
left-closed: breaks defined by c(2, 5, 10, 15, 20)
will be mapped as [2 - 5[, [5 - 10[, [10 - 15[, [15 - 20].
mtq <- mf_get_mtq()
mf_get_breaks(x = mtq$MED, nbreaks = 6, breaks = "quantile")
Run the code above in your browser using DataLab