lmestDecoding {LMest}

Perform local and global decoding

Description

Function that performs local and global decoding (Viterbi algorithm) from the output of lmest, lmestCont, and lmestMixed.

Usage

lmestDecoding(est, sequence = NULL, fort = TRUE, ...)
## S3 method for class 'LMbasic'
lmestDecoding(est, sequence = NULL,fort = TRUE, ...)
## S3 method for class 'LMmanifest'
lmestDecoding(est, sequence = NULL, fort = TRUE, ...)
## S3 method for class 'LMlatent'
lmestDecoding(est, sequence = NULL, fort = TRUE,...)
## S3 method for class 'LMbasiccont'
lmestDecoding(est, sequence = NULL, fort = TRUE,...)
## S3 method for class 'LMmixed'
lmestDecoding(est, sequence = NULL, fort = TRUE,...)

Arguments

est

an object obtained from a call to lmest, lmestCont, and lmestMixed

sequence

an integer vector indicating the units for the decoding. If NULL the whole observations are considered. (By default is set to NULL)

fort

to use fortran routines when possible

...

further arguments

Value

Ul

matrix of local decoded states corresponding to each row of Y

Ug

matrix of global decoded states corresponding to each row of Y

Author(s)

Francesco Bartolucci, Silvia Pandolfi, Fulvia Pennoni, Alessio Farcomeni, Alessio Serafini

References

Viterbi A. (1967) Error Bounds for Convolutional Codes and an Asymptotically Optimum Decoding Algorithm. IEEE Transactions on Information Theory, 13, 260-269.

Juan B., Rabiner L. (1991) Hidden Markov Models for Speech Recognition. Technometrics, 33, 251-272.

Examples

# Decoding for basic LM model

data("data_drug")
long <- data_drug[,-6]-1
long <- data.frame(id = 1:nrow(long),long)
long <- reshape(long,direction = "long",
                idvar = "id",
                varying = list(2:ncol(long)))

est <- lmest(index = c("id","time"),
             k = 3, 
             data = long,
             weights = data_drug[,6], 
             modBasic = 1)

# Decoding for a single sequence

out1 <- lmestDecoding(est, sequence = 1)

out2 <- lmestDecoding(est, sequence = 1:4)

# Decoding for all sequences

out3 <- lmestDecoding(est)

## Not run: 
# Decoding for LM model  with covariates on the initial and transition probabilities

data("data_SRHS_long")

SRHS <- data_SRHS_long[1:2400,]

# Categories rescaled to vary from 0 (“poor”) to 4 (“excellent”)

SRHS$srhs <- 5 - SRHS$srhs

est2 <- lmest(responsesFormula = srhs ~ NULL,
              latentFormula =  ~
              I(gender - 1) +
              I( 0 + (race == 2) + (race == 3)) +
              I(0 + (education == 4)) +
              I(0 + (education == 5)) +
              I(age - 50) + I((age-50)^2/100),
              index = c("id","t"),
              data = SRHS,
              k = 2,
              paramLatent = "difflogit",
              output = TRUE)

# Decoding for a single sequence

out3 <- lmestDecoding(est2, sequence = 1)

# Decoding for the first three sequences

out4 <- lmestDecoding(est2, sequence = 1:3)

# Decoding for all sequences

out5 <- lmestDecoding(est2)

## End(Not run)