Recoding Rules
Many recoding methods take a partition as a first parameter and a value of the same type (Numerical
or Categorical
) as a second one. They retrieve the part of the partition containing the value,
and output an index, which may be
- an integer value ranging from 1 to S, where S is the partition size,
- or an identifier (like
I1
), which is a genericCategorical
value, - or a label (like
]-∞; 2]
), which is a comprehensibleCategorical
value.
InInterval
Returns 1
if the input value belongs to the interval, 0
otherwise. The interval bounds must
contain exactly two bounds, for intervals of type ]lowerBound; upperBound]
. For left-open or
right-open intervals, use comparison derivation rules, such as G
,
GE
, L
or
LE
.
InGroup
Returns 1 if the input value belongs to the value group, 0 otherwise.
CellIndex
Computes the Numerical
cell index of a list of values given a data grid. The list of values are
either numerical Numerical
or Categorical
according to the data grid input partition types.
CellId
Computes the Categorical
cell identifier of a list of values given a data grid.
CellLabel
Computes the cell label of a list of values given a data grid.
ValueIndexDG
Computes the Numerical
index of a value given a univariate data grid.
PartIndexAt
Computes the Numerical
part index of a value given a data grid and an index of dimension in the
data grid. The index must be between 1 and the number of dimension in the data grid, and the value
of the type (Numerical or Categorical) relative to the given dimension of the data grid.
PartIdAt
Computes the categorical part identifier of a value given a data grid and an index of dimension in the data grid.
ValueRank
Computes the average normalized rank of a numerical value given a univariate numerical data grid.
InverseValueRank
Computes the average value related to a normalized rank given a univariate numerical data grid.
DataGridStats
Computes statistics (conditional probabilities) for a list of values given a data grid. The number of input values correspond of the inputs variables, which might be inferior to the number of dimensions in that data grid. The remaining dimensions in the data grid correspond to the output variables.
SourceConditionalInfo
Computes the source conditional info (negative log of the conditional probability) for the input cell related to the data grids stats and for the index of the target cell of the data grid given as a parameter (index starts at 1).
IntervalId
Computes the Categorical
part identifier of a Numerical
value given a partition into intervals.
ValueId
Computes the Categorical
part identifier of a Numerical
value given a partition into a set of
numerical values.
GroupId
Computes the Categorical
part identifier of a Categorical
value given a partition into a set of
groups of categorical values.
ValueIdC
Computes the Categorical
part identifier of a Categorical
value given a partition into a set of
categorical values. If value is not found in the value set, returns the part identifier of the
special value *
if it is defined, 1 otherwise.
IntervalIndex
Computes the Numerical
part index of a Numerical
value given a partition into intervals.
ValueIndex
Computes the Numerical
part index of a Numerical
value given a partition of numerical value
sets.
GroupIndex
Computes the Numerical
part index of a Categorical
value given a partition of groups of
categorical values.
ValueIndexC
Computes the Numerical
part index of a Categorical
value given a partition into a set of
Categorical
values. If the value is not found in the value set, it returns the index of the
special value *
if it is defined, 1 otherwise.