Allow shifting the start date of the yearly seasonality plot

R/R/prophet.R

@@ -990,13 +990,17 @@ plot.prophet <- function(x, fcst, uncertainty = TRUE, plot_cap = TRUE,
 #' @param weekly_start Integer specifying the start day of the weekly
 #'  seasonality plot. 0 (default) starts the week on Sunday. 1 shifts by 1 day
 #'  to Monday, and so on.
+#' @param yearly_start Integer specifying the start day of the yearly
+#'  seasonality plot. 0 (default) starts the year on Jan 1. 1 shifts by 1 day
+#'  to Jan 2, and so on.
 #'
 #' @return Invisibly return a list containing the plotted ggplot objects
 #'
 #' @export
 #' @importFrom dplyr "%>%"
-prophet_plot_components <- function(m, fcst, uncertainty = TRUE,
-                                    plot_cap = TRUE, weekly_start = 0) {
+prophet_plot_components <- function(
+    m, fcst, uncertainty = TRUE, plot_cap = TRUE, weekly_start = 0,
+    yearly_start = 0) {
   df <- df_for_plotting(m, fcst)
   # Plot the trend
   panels <- list(plot_trend(df, uncertainty, plot_cap))
@@ -1010,7 +1014,7 @@ prophet_plot_components <- function(m, fcst, uncertainty = TRUE,
   }
   # Plot yearly seasonality, if present
   if ("yearly" %in% colnames(df)) {
-    panels[[length(panels) + 1]] <- plot_yearly(m, uncertainty)
+    panels[[length(panels) + 1]] <- plot_yearly(m, uncertainty, yearly_start)
   }
   # Make the plot.
   grid::grid.newpage()
@@ -1119,12 +1123,16 @@ plot_weekly <- function(m, uncertainty = TRUE, weekly_start = 0) {
 #'
 #' @param m Prophet model object.
 #' @param uncertainty Boolean to plot uncertainty intervals.
+#' @param yearly_start Integer specifying the start day of the yearly
+#'  seasonality plot. 0 (default) starts the year on Jan 1. 1 shifts by 1 day
+#'  to Jan 2, and so on.
 #'
 #' @return A ggplot2 plot.
-plot_yearly <- function(m, uncertainty = TRUE) {
+plot_yearly <- function(m, uncertainty = TRUE, yearly_start = 0) {
   # Compute yearly seasonality for a Jan 1 - Dec 31 sequence of dates.
   df.y <- data.frame(
-    ds=seq.Date(zoo::as.Date('2017-01-01'), by='d', length.out=365), cap=1.)
+    ds=seq.Date(zoo::as.Date('2017-01-01'), by='d', length.out=365) +
+    yearly_start, cap=1.)
   df.y <- setup_dataframe(m, df.y)$df
   seas <- predict_seasonal_components(m, df.y)
   seas$ds <- df.y$ds

R/man/plot_yearly.Rd

@@ -4,12 +4,16 @@
 \alias{plot_yearly}
 \title{Plot the yearly component of the forecast.}
 \usage{
-plot_yearly(m, uncertainty = TRUE)
+plot_yearly(m, uncertainty = TRUE, yearly_start = 0)
 }
 \arguments{
 \item{m}{Prophet model object.}
 
 \item{uncertainty}{Boolean to plot uncertainty intervals.}
+
+\item{yearly_start}{Integer specifying the start day of the yearly
+seasonality plot. 0 (default) starts the year on Jan 1. 1 shifts by 1 day
+to Jan 2, and so on.}
 }
 \value{
 A ggplot2 plot.

R/man/prophet_plot_components.Rd

@@ -7,7 +7,7 @@ Prints a ggplot2 with panels for trend, weekly and yearly seasonalities if
 present, and holidays if present.}
 \usage{
 prophet_plot_components(m, fcst, uncertainty = TRUE, plot_cap = TRUE,
-  weekly_start = 0)
+  weekly_start = 0, yearly_start = 0)
 }
 \arguments{
 \item{m}{Prophet object.}
@@ -23,6 +23,10 @@ figure, if available.}
 \item{weekly_start}{Integer specifying the start day of the weekly
 seasonality plot. 0 (default) starts the week on Sunday. 1 shifts by 1 day
 to Monday, and so on.}
+
+\item{yearly_start}{Integer specifying the start day of the yearly
+seasonality plot. 0 (default) starts the year on Jan 1. 1 shifts by 1 day
+to Jan 2, and so on.}
 }
 \value{
 Invisibly return a list containing the plotted ggplot objects

python/fbprophet/forecaster.py

@@ -915,7 +915,7 @@ class Prophet(object):
         return fig
 
     def plot_components(self, fcst, uncertainty=True, plot_cap=True,
-                        weekly_start=0):
+                        weekly_start=0, yearly_start=0):
         """Plot the Prophet forecast components.
 
         Will plot whichever are available of: trend, holidays, weekly
@@ -930,6 +930,9 @@ class Prophet(object):
         weekly_start: Optional int specifying the start day of the weekly
             seasonality plot. 0 (default) starts the week on Sunday. 1 shifts
             by 1 day to Monday, and so on.
+        yearly_start: Optional int specifying the start day of the yearly
+            seasonality plot. 0 (default) starts the year on Jan 1. 1 shifts
+            by 1 day to Jan 2, and so on.
 
         Returns
         -------
@@ -958,7 +961,8 @@ class Prophet(object):
                 artists += self.plot_weekly(ax=ax, uncertainty=uncertainty,
                                             weekly_start=weekly_start)
             elif plot == 'yearly':
-                artists += self.plot_yearly(ax=ax, uncertainty=uncertainty)
+                artists += self.plot_yearly(ax=ax, uncertainty=uncertainty,
+                                            yearly_start=yearly_start)
 
         fig.tight_layout()
         return artists
@@ -1072,7 +1076,7 @@ class Prophet(object):
         ax.set_ylabel('weekly')
         return artists
 
-    def plot_yearly(self, ax=None, uncertainty=True):
+    def plot_yearly(self, ax=None, uncertainty=True, yearly_start=0):
         """Plot the yearly component of the forecast.
 
         Parameters
@@ -1080,6 +1084,9 @@ class Prophet(object):
         ax: Optional matplotlib Axes to plot on. One will be created if
             this is not provided.
         uncertainty: Optional boolean to plot uncertainty intervals.
+        yearly_start: Optional int specifying the start day of the yearly
+            seasonality plot. 0 (default) starts the year on Jan 1. 1 shifts
+            by 1 day to Jan 2, and so on.
 
         Returns
         -------
@@ -1091,7 +1098,8 @@ class Prophet(object):
             ax = fig.add_subplot(111)
         # Compute yearly seasonality for a Jan 1 - Dec 31 sequence of dates.
         df_y = pd.DataFrame(
-            {'ds': pd.date_range(start='2017-01-01', periods=365), 'cap': 1.})
+            {'ds': pd.date_range(start='2017-01-01', periods=365) +
+             pd.Timedelta(days=yearly_start), 'cap': 1.})
         df_y = self.setup_dataframe(df_y)
         seas = self.predict_seasonal_components(df_y)
         artists += ax.plot(df_y['ds'], seas['yearly'], ls='-',