plotTSGrouped: Multiple time series plot

Description Usage Arguments See Also Examples

View source: R/plotTSGrouped.R

Description

This function draws a time series of the values of variable 'Yvar' with the time variable 'Xvar' on the x-axis. The categorical variable that specify the group of the observations for which there will be one time series each.
Expects aggregated data.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
plotTSGrouped(
  .data,
  xvar = "",
  xlabel = "",
  yvar = "",
  ylabel = "",
  group = "",
  fill_color = EcdcColors(col_scale = "qual", n = length(unique(.data[[group]]))),
  log10_scale = FALSE,
  xvar_format = "%Y",
  xvar_breaks = "1 year"
)

Arguments

.data

dataframe containing the variables to plot

xvar

character string, name of the time variable (expects date format) to plot on the x-axis in quotes

xlabel

character string, label of the x axis

yvar

character string, name of the numerical variable to plot on the y-axis in quotes

ylabel

character string, label of the y axis

group

character string, name of the grouping variable in quotes, e.g. gender.

fill_color

character string, hexadecimal colour to use in the graph; (default to ECDC green "#65B32E", see EcdcColors(col_scale = "qual", n = length(unique(.data[[group]])))

log10_scale

boolean, TRUE if y-axis should be log scale (default FALSE ,see ggplot2::scale_y_log10)

xvar_format

character string, time format to use to plot the x-axis ("%Y" for yearly labels or "%b %Y" for monthly labels)

xvar_breaks

character string, time unit to use to plot the x-axis between breaks ("1 year" or "1 month", see ggplot2::scale_x_date(date_breaks = ... ))

See Also

Internal function: EcdcColors
Required Packages: ggplot2

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
# --- Create dummy data
mydat <- data.frame(TimeCode = seq(as.Date("2008/1/1"), as.Date("2017/1/1"), "years"),
                    YValue = sample(1:79/10, 20),
                    YLabel = rep(c("Acute", "Chronic"), each = 10))

# --- Plot the dummy data
plotTSGrouped(mydat,
              xvar = "TimeCode",
              xlabel = "Year",
              yvar = "YValue",
              ylabel = "Rate per 100 000 population",
              group = "YLabel",
              log10_scale = TRUE,
              xvar_format = "%Y",
              xvar_breaks = "1 year")

# --- Create dummy data
mydat <- data.frame(TimeCode = rep(seq(as.Date("2008/1/1"), as.Date("2017/1/1"), "years"), 5),
                    YValue = c(sample(1:50/10, 10),
                               sample(1:100/10, 10),
                               sample(1:300/10, 10),
                               sample(1:400/10, 10),
                               sample(1:500/10, 10)),
                    YLabel = rep(c("United Kingdom",
                                   "France",
                                   "Spain",
                                   "Netherlands",
                                   "Belgium"), each = 10))

# --- Plot the dummy data
plotTSGrouped(mydat,
              xvar = "TimeCode",
              xlabel = "Year",
              yvar = "YValue",
              ylabel = "Rate per 100 000 population",
              group = "YLabel",
              log10_scale = FALSE,
              xvar_format = "%Y",
              xvar_breaks = "1 year")

EpiReport documentation built on July 5, 2021, 5:06 p.m.