This release migrates graphics functionality to {ggtime}. The commonly used graphics functions are currently exported with a soft deprecation message. To use the gg_*() time series plot helper functions please include library(ggtime) in your code.
After a (very) gradual deprecation process, the {ggtime} dependency will be removed and the graphics functions will stop being re-exported. This deprecation process is planned to span approximately 2 years, after which explicitly using {ggtime} will be required.
Compatibility release for upcoming ggplot2 4.0.0 release.
gg_tsresiduals() now supports the plot_type argument to customise the
third plot, much like gg_tsdisplay().All ggplot2 functionality is now soft deprecated, and is being moved to the
new ggtime package. This deprecation process will be very gradual, although
the intention is for these functions to eventually be removed from feasts.
This is a design change to focus feasts on feature extraction and statistics
for time series, and have all ggplot2 functionality in a dedicated package
gg_season() not working with daily data showing seasonality > 1 week.gg_irf() for plotting impulse responses (typically obtained from using
IRF() with fable models).cointegration_johansen() and
cointegration_phillips_ouliaris() from urca.gg_season() not wrapping across facet_period argument correctly.Minor patch to resolve CRAN check issues with ggplot2 v3.5.0 breaking changes.
gg_season() breaks issue with ggplot2 v3.5.0Minor patch to resolve CRAN check issues with S3 method consistency.
tapered argument to ACF() and PACF() for producing banded and
tapered estimates of autocovariance (#1).gg_season() now allows seasonal period identifying labels to be nudged and
repelled with the labels_repel, labels_left_nudge, and
labels_right_nudge arguments.gg_season() behaviour of max_col has been restored, where colours aren't
used if the number of subseries to be coloured exceeds this value. The default
has changed to Inf since this function now supports continuous colour
guides. A new argument max_col_discrete has been added to control the
threshold for showing discrete and continuous colour guides (#150).guerrero() method to maintain a consistent subseries length by
removing the first few observations of needed. This more closely matches
the described method, and the implementation in the forecast package.grid.draw() method for ensemble graphics (gg_tsdisplay() and
gg_tsresiduals()). This allows use of ggsave() with these plots (#149).generate(<STL>) returning $.sim as a num [1:n(1d)] instead of
num [1:72] (fable/#336).gg_season() incorrectly grouping some seasonal subseries.CCF() now matches stats::ccf() x and y arguments (#144).Minor release for compatibility with an upcoming ggplot2 release. This release contains a few bug fixes and improvements to existing functionality.
gg_tsresiduals() function now allows the type of plotted residual to be
controlled via the type argument.STL() decompositions. For data with
a single seasonal pattern, the window has changed from 13 to 11. This change
is based on results from simulation experiments.seasonal::seas() defaults were not being used in
X_13ARIMA_SEATS() when defaults = "seasonal" (#130).gg_subseries() on data with spaces in the index column
name (#136).... in ACF(), PACF(), and CCF() with y (and x
for CCF()) arguments. This change should not affect the code for most users,
but is important for the eventual passing of ... to acf(), pacf() and
ccf() in a future version (#124).Small patch to fix check issues on Solaris, and to resolve components() for
automatically selected transformations in X_13ARIMA_SEATS().
X_13ARIMA_SEATS() decomposition method. This is a complete wrapper of
the X-13ARIMA-SEATS developed by the U.S. Census Bureau, implemented via
the seasonal::seas() function. The defaults match what is used in the
seasonal pacakge, however these defaults can be removed (giving an empty
default model) by setting defaults="none".X_13ARIMA_SEATS() method officially deprecates (supersedes) the
X11() and SEATS() models which were previously not exported (#66).generate() method for STL() decompositions. The method uses a block
bootstrap method to sample from the residuals.fitted() and residuals() methods for STL() decompositions.guerrero() default lower bound for Box-Cox lambda selection to from
-1 to -0.9. A transformation parameter of -1 typically results from data which
should not be transformed with a Box-Cox transformation, and can result in
very inaccurate forecasts if such a strong and inappropriate transformation is
used.A minor release to fix check issues introduced by changes in an upstream dependency.
gg_season() labels are low aligned outward (#115).gg_season() and
gg_subseries() (#117).gg_season()gg_lag() facets are now displayed with a 1:1 aspect ratio.n_flat_spots() function has been renamed to longest_flat_spot() to
more accurately describe the feature.gg_season() and ggsubseries() date structure improvements.n_flat_spots() return name is now "longest_flat_spot" to better describe
the feature.gg_tsdisplay() erroring when the spec.ar
order is chosen to be 0.CCF() lag being spaced by multiples of the data's frequency.gg_season() and gg_subseries() (#107).View() not working on ACF(), PACF() and CCF() outputs.Minor patch to resolve upstream check issues introduced by dplyr v1.0.0 and tsibble v0.9.0.
polar = TRUE in
gg_season().ACF().feat_spectral() to use stats::spec.ar() instead of
ForeCA::spectral_entropy(). Note that the feature value will be slightly
different due to use of a different spectral estimator, and the fix of a
bug in ForeCA.feat_stl().gg_lag() have been reversed for consistency with stats::lag.plot().feat_intermittent()gg_tsdisplay() not working with plotting expressions of data.gg_subseries() erroring when certain column names are used (#95).STL() specials.var_tiled_var() no longer includes partial tile windows in the computation.feat_stl().components().
For example, tourism %>% STL(Trips) is now tourism %>% model(STL(Trips)) %>% components().
This change allows for more flexible decomposition specifications, and better interfaces for decomposition modelling.feat_spectral() not showing results.ACF(), PACF() and CCF() for tidyr change.gg_tsdisplay() will no longer fail on non-seasonal data with missing values. The last plot will instead show a PACF in this case (#76)stat_arch_lm() (#85)gg_season, gg_subseries, gg_lag, gg_tsdisplay, gg_tsresiduals, gg_arma.ACF, PACF, CCF, and autoplot.tbl_cffabletools::features(): feat_stl, feat_acf, feat_pacf, guerrero, unitroot_kpss, unitroot_pp, unitroot_ndiffs, unitroot_nsdiffs, box_pierce, ljung_box, var_tiled_var, var_tiled_mean, shift_level_max, shift_var_max, shift_kl_max, feat_spectral, n_crossing_points, n_flat_spots, coef_hurst, stat_arch_lmclassical_decomposition, STL