Skip to contents

Merge event data together

merge_events.Rd

Merge events data set produced by process_events() with any extra events that needed to be included by an external source, such as the Actiwatch and sleep diary.

Usage

merge_events(
  data,
  add_events,
  start_time = NA,
  end_time = NA,
  off_times = NULL,
  good_days = 1:9,
  remove_days = FALSE
)

Arguments

data

A process.data.events object produced by process_events()

add_events

A data.frame of events to merge with data that has the same variable names as data.

start_time

A date-time (POSIXct) value that denotes first date and time that should be recorded in merged data set. Default is NA to indicate that the merge data set does not need to be subsetted.

end_time

A date-time (POSIXct) value that denotes last date and time that should be recorded in merged data set. Default is NA to indicate that the merge data set does not need to be subsetted.

off_times

A list containing two objects of equal length denoting the beginning and ending times of when the monitors were removed. For the Pregnancy 24/7 study, the windows_monitor() function was used to find these times. Default is NULL, which indicates that is no monitor off times.

good_days

An integer vector denoting the wear days that have valid data. The default is 1:9, as for the Pregnancy 24/7 study the wear period last nine days.

remove_days

A logical denoting if merge data set should be subsetted by the good_days. Default is FALSE.

Value

A merge.data.events object, which is a data.frame with all of the variables in data plus the event variable.

Details

After data and add_events have been merged together, all missing values created by merging the two data set together are filled in by their appropriate values. For the looping variables, usually, the last value is carried forwarded. As for the activity variables, it depends on how the original variable was created as for how the missing values are filled in.

If there are monitor off times, the looping variables are replaced with the value of 99 to indicate a special type of missing values related to monitor wear.

There are two more objectives of this function.

  1. Round the time variable to the nearest second with lubridate::round_date() and

  2. create a new variable called event that denotes the event in recorded.

See also

lubridate::round_date(), tidyr::fill()

Examples

# See process_data() for an example of how this function is used.