Data Structure

• StatLink 1: http://dx.doi.org/10.1787/888933437487
• StatLink 2: http://dx.doi.org/10.1787/888933437497

## [1] "can11.tsv" "can12.tsv" "can21.tsv" "can22.tsv"

date	mining	other	total
2013-01-01	-2.850581	1.554630	1.187306
2013-02-01	1.817405	1.772702	1.776345
2013-03-01	2.894831	1.879273	1.961509

date	mining	other
2010-01-01	12.58940	1.276840
2010-04-01	33.89071	5.842328
2010-07-01	40.60906	6.814514

date	vancouver	toronto	rest
2010-01-01	85.05708	78.44315	88.95637
2010-02-01	85.43009	78.75323	89.03557
2010-03-01	85.98395	78.69509	89.26357

date	ontario	other	total
2010-01-01	7.163440	13.00803	9.857602
2010-04-01	8.132658	18.30270	13.314303
2010-07-01	10.480245	12.00953	11.249248

Static Chart 1

FAME Output

ggplot2 Output

• ggplot2: Specify a secondary axis using sec_axis
• alternative: plotting independent series: Dual axis in ggplot2

ggplotly

Issues: R plotly currently does not draw the secondary y-axis defined using the sec.axis parameter. Also, modifications to axis title positions and legends are not reflected properly and multi-line titles overlap with the plotly mode bar

latticeExtra Double Y-Scales

Static Chart 2

FAME Output

lattice Output

• x-axis subticks: https://github.com/Rapporter/pander/blob/master/R/graph.R and Sarkar, 2008, Lattice: Multivariate Data Visualization with R, Springer, pp.146f.
• missing: right y-axis label, automatic left y-axis label position, quarterly minor ticks, legend inside plot area in doubleYScale using corner([0;1],[0;1]), flashes outside legend, legend line colors on left-hand side
• todo: expand x-axis, increase relative font size axis tick labels, increase visibility of grid lines using panel.grid()

Dygraphs

• xts: Plotting Time Series in R using Yahoo Finance data

Todo

• fix JS: keep axis ticks when drawYAxis = FALSE, keep horizontal lines when drawGrid = FALSE using drawXGrid and drawYGrid from version 1.1.0 (current version 2.0.0)

Plotly

• plot.ly: Multiple Axes in R
• replace plotly icon in hover modebar (see http://rdata.work/technical/2015/07/26/javascript.html, “plotly.js”)
• the modified plotly.js library can be found at https://s3.eu-central-1.amazonaws.com/cdn-plotly/plotly-custom.min.js
• export SVG from plotly: plotly::export() function allows exporting an SVG vector image using the RSelenium package to establish a connection to a web browser

Plotly Static Image Export (SVG)

fig_path <- "assets/img"
file_name <- "plotly-multiple-y"
svg_file <- paste0(file_name, ".svg")
svg_tofile <- file.path(fig_path, svg_file)
unlink(svg_tofile)
## chrome options
eCaps <- list(
  chromeOptions =
    list(prefs = list(
      "profile.default_content_settings.popups" = 0L,
      "download.prompt_for_download" = FALSE,
      "download.default_directory" = file.path(getwd(), fig_path)
      )))

## create plot
suppressMessages(
  rD <- RSelenium::rsDriver(port = 4590L,
                            browser = "chrome",
                            geckover = NULL,
                            iedrver = NULL,
                            phantomver = NULL,
                            extraCapabilities = eCaps,
                            verbose = FALSE)
)

p$layout$height <- 480                  # match interactive plot height
export(p = p, file = svg_file, selenium = rD)
Sys.sleep(1)
rDstop <- rD[["server"]]$stop()
knitr::include_graphics(svg_tofile)

## convert to print formats
rsvg_pdf(svg_tofile, file.path(fig_path, paste0(file_name, ".pdf")))
rsvg_ps(svg_tofile, file.path(fig_path, paste0(file_name, ".eps")))

• PDF file
• EPS file

ggiraph

	date	industry	value	value_scaled	indlabel
1	2013-01-01	mining	-2.850581	-0.7126454	Mining, quarrying, oil and gas extraction –>
49	2013-01-01	other	1.554630	1.5546297	<– Other
97	2013-01-01	total	1.187306	1.1873061	<– All industries