setBaseCat#
Purpose#
Assign the label basecase to be the base case for the categorical variable specified by columns .
Format#
- x_meta = setBaseCat(x, basecase[, columns])#
- Parameters:
x (NxK matrix) – data.
basecase (Mx1 string array) – category to be set to base case.
columns (Mx1 scalar or string) – Optional argument, indicates columns of categorical variables to set base case for. Default = all columns.
- Returns:
x_meta (NxK dataframe) – contains data with categorical base cases set to the categories specified in basecase for the variables in columns .
Remarks#
Use reordercatlabels()
to change the order of all category labels.
Examples#
// Load yarn data file
fname = getGAUSSHome("examples/yarn.xlsx");
yarn = loadd(fname, "cat(yarn_length) + cycles");
// Get categorical labels for 'yarn_length'
print "Original categorical labels:";
getColLabels(yarn, "yarn_length");
// Change base case to 'low'
yarn = setBaseCat(yarn, "low", "yarn_length");
print "";
print "Updated categorical labels:";
getColLabels(yarn, "yarn_length");
The above code will print out:
Original categorical labels:
high
low
med
0.0000000
1.0000000
2.0000000
Updated categorical labels:
low
high
med
0.0000000
1.0000000
2.0000000
See also
Functions dftype()
, getColLabels()
, setColLabels()