Tabulate observed transitions between states over successive observations, by from-state, to-state and (optionally) time interval length and covariate values.
Usage
statetable(
data,
state = "state",
subject = "subject",
time = "time",
covariates = NULL,
time_groups = 1,
format = "wide"
)Arguments
- data
Data frame giving the observed data.
- state
Character string naming the observed state variable in the data. This variable must either be an integer in 1,2,...,K, where K is the number of states, or a factor with these integers as level labels. If omitted, this is assumed to be
"state".- subject
Character string naming the individual ID variable in the data. If omitted, this is assumed to be
"subject".- time
Character string naming the observation time variable in the data. If omitted, this is assumed to be
"time".- covariates
Vector of names of covariates to summarise counts by.
- time_groups
Number of groups to summarise the time intervals by. The transitions are categorised into groups according to equally-spaced quantiles of the time interval length.
- format
"long"to return one row pertostate(a pure "tidy data" format) or"wide"to return one column pertostate(likestatetable.msmin msm).
Value
A data frame with columns fromstate, timelag and
n (count of transitions), and column or columns for tostate.
Details
This is like the function statetable.msm in msm, except that it
uses msmbayes syntax for specifying the data, it summarises
the length of the time intervals between successive observations,
and it returns a tidy data frame.
Warning: it is not appropriate to choose the transition structure
(the Q argument to msmbayes()) on the basis of this summary.
statetable counts transitions over a time interval, whereas
Q indicates which instantaneous transitions are possible.
The structures will not be the same. For example, in a model with
instananeous transitions from mild to moderate illness, and moderate
to severe, we might observe transitions from mild to severe over
an interval of 1 year (say), but the instantaneous transition from
mild to severe is impossible.
Note this is not fully tidy-friendly, as it will not work
if data is grouped using dplyr.