tsibble-package: tsibble: tidy temporal data frames and tools

Description

The tsibble package provides a data class of tbl_ts to represent tidy temporal-context data. A tsibble consists of a time index, key, and other measured variables in a data-centric format, which is built on top of the tibble.

Arguments

Index

The time indices are preserved as the essential data component of the tsibble, instead of implicit attribute (for example, the tsp attribute in a ts object). A few index classes, such as Date, POSIXct, and difftime, forms the basis of the tsibble, with new additions yearweek, yearmonth, and yearquarter representing year-week, year-month, and year-quarter respectively. Any arbitrary index class are also supported, including zoo::yearmth, zoo::yearqtr, and nanotime. For a tbl_ts of regular interval, a choice of index representation has to be made. For example, a monthly data should correspond to time index created by yearmonth or zoo::yearmth, instead of Date or POSIXct. Because months in a year ensures the regularity, 12 months every year. However, if using Date, a month contains days ranging from 28 to 31 days, which results in irregular time space. This is also applicable to year-week and year-quarter.

Since the tibble that underlies the tsibble only accepts a 1d atomic vector or a list, a tbl_ts doesn't accept POSIXlt and timeDate columns.

Key

Key variable(s) together with the index uniquely identifies each record. And the key also imposes the structure on a tsibble, which can be created via the id function as identifiers:

None: an implicit variable id() resulting a univariate time series.
A single variable: an explicit variable. For example, data(pedestrian) uses the id(Sensor) column as the key.
Nested variables: a nesting of one variable under another. For example, data(tourism) contains two geographical locations: Region and State. Region is the lower level than State in Australia; in other words, Region is nested into State, which naturally forms a hierarchy. A vertical bar (|) is used to describe this nesting relationship, and thus Region | State. Alternatively, a forward slash (/) expresses the equivalent hierarchy but in a reverse order, for example State / Region. In theory, nesting can involve infinite levels, so is tsibble.
Crossed variables: a crossing of one variable with another. For example, the geographical locations are crossed with the purpose of visiting (Purpose) in the data(tourism). A comma (,) is used to indicate this crossing relationship. Nested and crossed variables can be combined, such as data(tourism) using id(Region | State, Purpose).

These key variables describe the data structure, which will prove useful in data visualisation and statistical modelling.

Interval

The interval function returns the interval associated with the tsibble.

Regular: the value and its time unit including "nanosecond", "microsecond", "millisecond", "second", "minute", "hour", "day", "week", "month", "quarter", "year". An unrecognisable time interval is labelled as "unit".
Irregular: as_tsibble(regular = FALSE) gives the irregular tsibble. It is marked with !.
Unknown: if there is only one entry for each key variable, the interval cannot be determined (?).

An interval is obtained based on the corresponding index representation:

integer/numeric: either "unit" or "year"
yearquarter/yearqtr: "quarter"
yearmonth/yearmth: "month"
yearweek: "week"
Date: "day"
POSIXct: "hour", "minute", "second", "millisecond", "microsecond"
nanotime: "nanosecond"

Print options

The tsibble package fully utilises the print method from the tibble. Please refer to tibble::tibble-package to change display options.

Examples

Run this code

# NOT RUN {
# create a tsibble w/o a key ----
tsibble(
  date = as.Date("2017-01-01") + 0:9,
  value = rnorm(10)
)

# create a tsibble with one key ----
tsibble(
  qtr = rep(yearquarter("201001") + 0:9, 3),
  group = rep(c("x", "y", "z"), each = 10),
  value = rnorm(30),
  key = id(group)
)
# }

Run the code above in your browser using DataLab