transMap: Summarizes the type and location of transitions between...

In pnnl/qFeature: Extract Features from Continuous or Discrete Time Series

Description

Summarizes the type and location of transitions between discrete states in an integer vector

Usage

transMap(y, y.unique)

Arguments

y	The integer vector whose state transitions will be identified
y.unique	An integer vector containing the unique values of y

Details

The algorithm applies a discrete, 1-1 map, which we'll call m, to the vector y such that the first differences of m(y) uniquely identify the transitions between all the possible values of y. It creates the map, calculates the first differences of m(y), and produces a set of transition labels that indicate which value of m(Y) corresponds to each of the possible transitions among the unique states of y.

Value

A list with the following components: (also see example below)

diffMap	The first differences of m(y) which identify the position and type of state transition in y
from	The 'from' state labels
to	The 'to' state labels
transValue	The value that diffMap will take when y transitions from the state label in from to the corresponding state label in to

Author(s)

Landon Sego

See Also

This function is called in discFeatures

Examples

transMap(c(1,1,1,2,2,1,1,3,3,2,2), 1:3)

# These are the mapped values which signal when
# a transition took place from one state to another
# and they uniquely identify which type of transistion
# it was.  Their values can be decoded using 'transValue'
# below.  A value of 0 indicates that no transition occured
# at that index.
# $diffMap
#  [1]  0  0  0 -1  0  1  0 -3  0  2  0

# The 'from' state labels
# $from
# [1] 1 1 2 2 3 3

# The 'to' state labels
# $to
# [1] 2 3 1 3 1 2

# The transistion value key.  e.g., -1 corresponds to a
# transition from state 1 to state 2.  -3 corresponds
# to a transition from state 1 to state 3.  etc.
# $transValue
# [1] -1 -3  1 -2  3  2

pnnl/qFeature documentation built on May 25, 2019, 10:22 a.m.