Skip to contents

FinanceGraphs-Customization

Source: vignettes/FinanceGraphs-customization.Rmd
FinanceGraphs-customization.Rmd

Overview of Aesthetics Organization

Managing a consistent look across graphs is not easy, as there are so many parameters that are possible to change. ggplot2 does a great job allowing every detail to be customized, especially with the use of themes. However, adding all those customizations are burdensome, and ad-hoc changes to them can involve a great deal of programming. Ideally, we would like to keep our favorite looks in one place with an easy ability to change just a few pieces in an ad-hoc manner.

The functions in the package attempt to ease that burden with a middle layer of named aesthetic groups. Aesthetics can be managed just like any other data, so internal to the package is a data.table with the following character columns, plus a few others:

Column Description
Category An “aesthetic set” key grouping together various subparts of a graph.
(Ordering) Variable Any string that can sorted to keep the aesthetics in a desired order
Type The type of aesthetic (e.g. color, size, symbol, etc)
Value The actual aesthetic to be used. If a number, then each function casts appropriately
used Where the function is used

Each FinanceGraphs function has internal aesthetic names (and defaults) that a user can modify, either persistently or temporarily. For example, the basic set of colors for categories is "lines", which are the line colors for fgts_dygraph() or the category colors for fg_scatplot.

fg_get_aes("lines", n_max=6)
#>    category variable  type     value const used                     helpstr
#> 1:    lines      D01 color     black        all Low cardinality line colors
#> 2:    lines      D02 color       red        all Low cardinality line colors
#> 3:    lines      D03 color      blue        all Low cardinality line colors
#> 4:    lines      D04 color darkgreen        all Low cardinality line colors
#> 5:    lines      D05 color    orange        all Low cardinality line colors
#> 6:    lines      D06 color      gray        all Low cardinality line colors

Changing the look of graphs is as easy as changing the values in that internal dataset. In some cases (e.g. in fg_scatplot()) a user can make an entire new aesthetic set with a new (unique) name and directly specify it in the function call. The package keeps those changes persistently (by default), so each users’ preferences should only need to be specified once.

Getting current aesthetics

There are two ways to get the aesthetic sets for every graph in the package:

  • Call the function fg_print_aes_list() with the name of a function as an argument
  • Call the function fg_verbose() to turn on logging of what aesthetic sets are called for any given graph generated.

For example, here are the a few aesthetic sets for fg_eventStudy(). This function just produces a summary, so for example there are actually 14 different line colors in "lines".

print(fg_print_aes_list("fg_eventStudy"))
#> 
#> 
#> |category    |helpstr                     |default   |  N|
#> |:-----------|:---------------------------|:---------|--:|
#> |lines       |Low cardinality line colors |black     | 14|
#> |espath_fill |Fitline fill color          |#C2DF23FF |  1|
#> |espath_gp   |Brewer colors if used       |seq,BuGn  |  1|
#> |espath_line |fitline color               |#433E85FF |  1|
#> |espath_lm   |Fit line color              |black     |  1|
#> |espath_ls   |Line styles per event       |84        |  9|
#> |espath_x    |Scatterplot x color         |gray      |  1|
#> |espath_y    |Scatterplot y color         |darkgreen |  1|

Each set can have multiple rows, as in

fg_get_aes("espath_ls", n_max=6)
#>     category variable      type value const          used               helpstr
#> 1: espath_ls      L01 linestyle    84       fg_eventStudy Line styles per event
#> 2: espath_ls      L02 linestyle  8282       fg_eventStudy Line styles per event
#> 3: espath_ls      L03 linestyle  8484       fg_eventStudy Line styles per event
#> 4: espath_ls      L04 linestyle    22       fg_eventStudy Line styles per event
#> 5: espath_ls      L05 linestyle    24       fg_eventStudy Line styles per event
#> 6: espath_ls      L06 linestyle  4242       fg_eventStudy Line styles per event

These correspond directly to ggplot2 aesthetics, such as the linetypes in linetypes. Some other notes:

  • Each value is a character column, but is coerced as necessary in each function.
  • In the cases of sizes, the values are multiples of the function parameters psize and tsize. So, a text size value of "6" gets converted to tsize * 6 in each function.
  • The "variable" column tells the functions to use "L01" first, "L02" second, etc. As long as the key is sortable, consistent and correct values should be used.
  • In the case of category overload, most functions have “color switch” parameters beyond which colors are “brewed” scales which are infinitely (but not wisely) expandable. In this case, the value is "seq,<colorset>" which generates colors from a Colorbrewer color set.
  • Colors can be color names or Hex color codes.

Colors in particular may be examined using the fg_display_colors() function as in

Adding or customizing aesthetics

Any of the default aethetic sets can be customized across calls to the functions and invocations of the package using fg_update_aes() New aesthetics sets can also be added for those functions (e.g. fg_scatplot()) where different aesthetic sets can be specified at runtime.

Modifying or adding aesthetics sets is done by creating (or copying and editing) a data.frame obtained from fg_get_aes() As a simple example, suppose we have three related classes of assets, one of which we wish to highlight and the others are related, but less important. Here is how the default colors would look:

onedt <- function(offset,category) { data.table(x=seq(1,10),y=seq(1,10)-offset,id=rep(category,10))}
exampledta <- rbind(onedt(0,"A"),onedt(1,"B"),onedt(2,"C"))
fg_scatplot(exampledta,"y ~ x + color:id",title="default",psize=6)

Customizing default aesthetics

The procedure for changing those colors is as follows:

  1. Get an example data.frame using fg_get_aes()
  2. Modify the values you want to change in the value column
  3. Update the persistent package data using fg_update_aes() 
head(oldcolors <- fg_get_aes("lines"),3)
#>    category variable  type value const used                     helpstr
#> 1:    lines      D01 color black        all Low cardinality line colors
#> 2:    lines      D02 color   red        all Low cardinality line colors
#> 3:    lines      D03 color  blue        all Low cardinality line colors
oldcolors[c(1,2,3),"value"] <- alpha("blue", c(1,0.6,0.3))
# Note that we still keep "category" as "lines".  To add a new set, use a different name.
fg_update_aes( oldcolors )
#> Saved aesthetic updates to /home/runner/.cache/R/FinanceGraphs/fg_aes.RD
fg_display_colors("lines")

fg_scatplot(exampledta,"y ~ x + color:id",title="default",psize=6)

Adding new aesthetics

To create our own aesthetics, we use the same procedure, but adding our own "category":

oldcolors[c(1,2,3),"value"] <- alpha("red", c(1,0.6,0.3))
oldcolors[c(1,2,3),"category"] <- rep("MyNewColors",3)
fg_update_aes( oldcolors )
#> Saved aesthetic updates to /home/runner/.cache/R/FinanceGraphs/fg_aes.RD
fg_scatplot(exampledta,"y ~ x + color:id,MyNewColors",title="with new colors",psize=6)

Themes

Themes are the ggplot2 way of proscribing every single aesthetic detail in a graph. This package uses a default theme derived from theme_bw(), but it is quite easy to create or modify, and more importantly save, a custom theme for future use.

To do so, just call fg_replace_theme() as in the following example:

fg_replace_theme(theme_dark())
#> Saved Default Theme to /home/runner/.cache/R/FinanceGraphs/fg_theme.RD
fg_scatplot(exampledta,"y ~ x + color:id,MyNewColors",title="New Theme, New Colors",psize=6)

Persistence

Saved Aesthetics, themes and dates of interest persist

This package manages aesthetic changes for you by caching the current aesthetic sets, themes, and dates of interest in local files, which are then loaded on package invocation. If you don’t want save changes, then call fg_update_aes() and fg_replace_theme() with persist=FALSE parameters.

To reset all parameters back to the package defaults, run fg_reset_to_default_state()

fg_reset_to_default_state("all")
#> Removing dates file and reverting to defaults of package
#> Removing Aesthetics file and reverting to defaults of package
#> Removing User-made Themes and reverting to defaults of package
#> Removing cache Directory
#> fg_reset_to_default_state(all) completed
fg_scatplot(exampledta,"y ~ x + color:id",title="Back to old colors",psize=6)