TQ04-charting-with-tidyquant.R
Charting financial data using
ggplot2
The tidyquant
package includes charting tools to assist
users in developing quick visualizations in ggplot2
using
the grammar of graphics format and workflow. There are currently three
primary geometry (geom) categories and one coordinate manipulation
(coord) category within tidyquant
:
geom_barchart
and
geom_candlestick
.geom_ma
.geom_bbands
. The BBand moving average can be one of
the seven available in Moving Averages.coord
functions are available (coord_x_date
and
coord_x_datetime
), which prevent data loss when zooming in
on specific regions of a chart. This is important when using the moving
average and Bollinger band geoms.Load the tidyquant
package to get started.
# Loads tidyquant, xts, quantmod, TTR, and PerformanceAnalytics
library(tidyverse)
library(tidyquant)
The following stock data will be used for the examples. Use
tq_get
to get the stock prices.
# Use FANG data set
# Get AAPL and AMZN Stock Prices
AAPL <- tq_get("AAPL", get = "stock.prices", from = "2015-09-01", to = "2016-12-31")
AMZN <- tq_get("AMZN", get = "stock.prices", from = "2000-01-01", to = "2016-12-31")
TQ04-charting-with-tidyquant.R
The end
date parameter will be used when setting date
limits throughout the examples.
## [1] "2016-12-31"
TQ04-charting-with-tidyquant.R
The AAPL_range_60
will be used to adjust the zoom on the
plots.
aapl_range_60_tbl <- AAPL %>%
tail(60) %>%
summarise(
max_high = max(high),
min_low = min(low)
)
aapl_range_60_tbl
## # A tibble: 1 × 2
## max_high min_low
## <dbl> <dbl>
## 1 29.7 26.0
TQ04-charting-with-tidyquant.R
Financial charts provide visual cues to open, high, low, and close prices. The following chart geoms are available:
geom_barchart
geom_candlestick
Before we visualize bar charts and candlestick charts using the
tidyquant
geoms, let’s visualize stock prices with a simple
line chart to get a sense of the “grammar of graphics” workflow. This is
done using the geom_line
from the ggplot2
package. The workflow begins with the stock data, and uses the pipe
operator (%>%
) to send to the ggplot()
function.
The primary features controlling the chart are the aesthetic
arguments: these are used to add data to the chart by way of the
aes()
function. When added inside the ggplot()
function, the aesthetic arguments are available to all underlying
layers. Alternatively, the aesthetic arguments can be applied to each
geom individually, but typically this is minimized in practice because
it duplicates code. We set aesthetic arguments, x = date
and y = close
, to chart the closing price versus date. The
geom_line()
function inherits the aesthetic arguments from
the ggplot()
function and produces a line on the chart.
Labels are added separately using the labs()
function.
Thus, the chart is built from the ground up by starting with data and
progressively adding geoms, labels, coordinates / scales and other
attributes to create a the final chart. This is enables maximum
flexibility wherein the analyst can create very complex charts using the
“grammar of graphics”.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_line() +
labs(title = "AAPL Line Chart", y = "Closing Price", x = "") +
theme_tq()
TQ04-charting-with-tidyquant.R
Visualizing the bar chart is as simple as replacing
geom_line
with geom_barchart
in the ggplot
workflow. Because the bar chart uses open, high, low, and close prices
in the visualization, we need to specify these as part of the aesthetic
arguments, aes()
. We can do so internal to the geom or in
the ggplot()
function.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_barchart(aes(open = open, high = high, low = low, close = close)) +
labs(title = "AAPL Bar Chart", y = "Closing Price", x = "") +
theme_tq()
TQ04-charting-with-tidyquant.R
We zoom into specific sections using coord_x_date
, which
has xlim
and ylim
arguments specified as
c(start, end)
to focus on a specific region of the chart.
For xlim
, we’ll use lubridate
to convert a
character date to date class, and then subtract six weeks using the
weeks()
function. For ylim
we zoom in on
prices in the range from 100 to 120.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_barchart(aes(open = open, high = high, low = low, close = close)) +
labs(title = "AAPL Bar Chart",
subtitle = "Zoomed in using coord_x_date",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(end - weeks(6), end),
ylim = c(aapl_range_60_tbl$min_low, aapl_range_60_tbl$max_high)) +
theme_tq()
TQ04-charting-with-tidyquant.R
The colors can be modified using colour_up
and
colour_down
arguments, and parameters such as
size
can be used to control the appearance.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_barchart(aes(open = open, high = high, low = low, close = close),
colour_up = "darkgreen", colour_down = "darkred", size = 1) +
labs(title = "AAPL Bar Chart",
subtitle = "Zoomed in, Experimenting with Formatting",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(end - lubridate::weeks(6), end),
c(aapl_range_60_tbl$min_low, aapl_range_60_tbl$max_high)) +
theme_tq()
TQ04-charting-with-tidyquant.R
Creating a candlestick chart is very similar to the process with the
bar chart. Using geom_candlestick
, we can insert into the
ggplot
workflow.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close)) +
labs(title = "AAPL Candlestick Chart", y = "Closing Price", x = "") +
theme_tq()
TQ04-charting-with-tidyquant.R
We zoom into specific sections using coord_x_date
.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close)) +
labs(title = "AAPL Candlestick Chart",
subtitle = "Zoomed in using coord_x_date",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(end - weeks(6), end),
c(aapl_range_60_tbl$min_low, aapl_range_60_tbl$max_high)) +
theme_tq()
TQ04-charting-with-tidyquant.R
The colors can be modified using colour_up
and
colour_down
, which control the line color, and
fill_up
and fill_down
, which control the
rectangle fills.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close),
colour_up = "darkgreen", colour_down = "darkred",
fill_up = "darkgreen", fill_down = "darkred") +
labs(title = "AAPL Candlestick Chart",
subtitle = "Zoomed in, Experimenting with Formatting",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(end - weeks(6), end),
c(aapl_range_60_tbl$min_low, aapl_range_60_tbl$max_high)) +
theme_tq()
TQ04-charting-with-tidyquant.R
We can use facet_wrap
to visualize multiple stocks at
the same time. By adding a group
aesthetic in the main
ggplot()
function and combining with a
facet_wrap()
function at the end of the ggplot
workflow, all four “FANG” stocks can be viewed simultaneously. You may
notice an odd filter()
call before the call to
ggplot()
. I’ll discuss this next.
start <- end - weeks(6)
FANG %>%
dplyr::filter(date >= start - days(2 * 15)) %>%
ggplot(aes(x = date, y = close, group = symbol)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close)) +
labs(title = "FANG Candlestick Chart",
subtitle = "Experimenting with Multiple Stocks",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(start, end)) +
facet_wrap(~ symbol, ncol = 2, scale = "free_y") +
theme_tq()
TQ04-charting-with-tidyquant.R
A note about out-of-bounds data (or “clipping”), which is particularly important with faceting and charting moving averages:
The coord_x_date
coordinate function is designed to zoom
into specific sections of a chart without “clipping” data that is
outside of the view. This is in contrast to scale_x_date
,
which removes out-of-bounds data from the charting. Under normal
circumstances clipping is not a big deal (and is actually helpful for
scaling the y-axis), but with financial applications users want to chart
rolling/moving averages, lags, etc that depend on data outside of the
view port. Because of this need for out-of-bounds data, there is a
trade-off when charting: Too much out-of-bounds data distorts
the scale of the y-axis, and too little and we cannot get a moving
average. The optimal method is to include “just enough”
out-of-bounds data to get the chart we want. This is why below the FANG
data is filtered by date from double the number of moving-average days
(2 * n
) previous to the start date. This yields a nice
y-axis scale and still allows us to create a moving average line using
geom_ma
.
start <- end - weeks(6)
FANG %>%
dplyr::filter(date >= start - days(2 * 15)) %>%
ggplot(aes(x = date, y = close, group = symbol)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close)) +
geom_ma(ma_fun = SMA, n = 15, color = "darkblue", size = 1) +
labs(title = "FANG Candlestick Chart",
subtitle = "Experimenting with Multiple Stocks",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(start, end)) +
facet_wrap(~ symbol, ncol = 2, scale = "free_y") +
theme_tq()
TQ04-charting-with-tidyquant.R
Moving
averages are critical to evaluating time-series trends.
tidyquant
includes geoms to enable “rapid prototyping” to
quickly visualize signals using moving averages and Bollinger bands.
The following moving averages are available:
Moving averages are applied as an added layer to a chart with the
geom_ma
function. The geom is a wrapper for the underlying
moving average functions from the TTR
package:
SMA
, EMA
, WMA
, DEMA
,
ZLEMA
, VWMA
, and EVWMA
. Here’s
how to use the geom:
ma_fun
, that you want
to apply.ma_fun
. You can investigate the underlying function by
searching ?TTR::SMA
.aes(x = date, y = close)
. The volume-weighted functions
will require the volume
argument in the aes()
function.ggplot
workflow.Important Note: When zooming in on a section, use
coord_x_date
or coord_x_datetime
to prevent
out-of-bounds data loss. Do not use scale_x_date
, which
will affect the moving average calculation. Refer to Charting Multiple Securities.
We want to apply a SMA
, so we research the TTR function
and we see that it accepts, n
, the number of periods to
average over. We see that the aesthetics required are x
, a
date, and y
, a price. Since these are already in the main
ggplot()
function, we don’t need to add the aesthetics to
the geom. We apply the moving average geoms after the candlestick geom
to overlay the moving averages on top of the candlesticks. We add two
moving average calls, one for the 50-day and the other for the 200-day.
We add color = "red"
and linetype = 5
to
distinguish the 200-day from the 50-day.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close)) +
geom_ma(ma_fun = SMA, n = 50, linetype = 5, size = 1.25) +
geom_ma(ma_fun = SMA, n = 200, color = "red", size = 1.25) +
labs(title = "AAPL Candlestick Chart",
subtitle = "50 and 200-Day SMA",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(end - weeks(24), end),
c(aapl_range_60_tbl$min_low * 0.9, aapl_range_60_tbl$max_high)) +
theme_tq()
TQ04-charting-with-tidyquant.R
We want an EMA
, so we research the TTR function and we
see that it accepts, n
, the number of periods to average
over, wilder
a Boolean, and ratio
arguments.
We will use wilder = TRUE
and go with the default for the
ratio
arg. We see that the aesthetics required are
x
, a date, and y
, a price. Since these are
already in the main ggplot()
function, we don’t need to
modify the geom. We are ready to apply after the bar chart geom.
AAPL %>%
ggplot(aes(x = date, y = close)) +
geom_barchart(aes(open = open, high = high, low = low, close = close)) +
geom_ma(ma_fun = EMA, n = 50, wilder = TRUE, linetype = 5, size = 1.25) +
geom_ma(ma_fun = EMA, n = 200, wilder = TRUE, color = "red", size = 1.25) +
labs(title = "AAPL Bar Chart",
subtitle = "50 and 200-Day EMA",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(end - weeks(24), end),
c(aapl_range_60_tbl$min_low * 0.9, aapl_range_60_tbl$max_high)) +
theme_tq()
TQ04-charting-with-tidyquant.R
We’ll double up using a volume-weighted average (VWMA
)
and apply it to the FANG stocks at once. Since VWMA is a volume-weighted
function, we need to add volume as an aesthetic. Because we are viewing
multiple stocks, we add a group aesthetic setting it to the symbol
column which contains the FANG stock symbols. The facet wrap is added at
the end to create four charts instead of one overlayed chart.
start <- end - weeks(6)
FANG %>%
dplyr::filter(date >= start - days(2 * 50)) %>%
ggplot(aes(x = date, y = close, volume = volume, group = symbol)) +
geom_candlestick(aes(open = open, high = high, low = low, close = close)) +
geom_ma(ma_fun = VWMA, n = 15, wilder = TRUE, linetype = 5) +
geom_ma(ma_fun = VWMA, n = 50, wilder = TRUE, color = "red") +
labs(title = "FANG Bar Chart",
subtitle = "50 and 200-Day EMA, Experimenting with Multiple Stocks",
y = "Closing Price", x = "") +
coord_x_date(xlim = c(start, end)) +
facet_wrap(~ symbol, ncol = 2, scales = "free_y") +
theme_tq()