TQ03-scaling-and-modeling-with-tidyquant.R
Designed for the data science workflow of the
tidyverse
The greatest benefit to tidyquant
is the ability to
apply the data science workflow to easily model and scale your financial
analysis as described in R for
Data Science. Scaling is the process of creating an analysis
for one asset and then extending it to multiple groups. This idea of
scaling is incredibly useful to financial analysts because typically one
wants to compare many assets to make informed decisions. Fortunately,
the tidyquant
package integrates with the
tidyverse
making scaling super simple!
All tidyquant
functions return data in the
tibble
(tidy data frame) format, which allows for
interaction within the tidyverse
. This means we can:
%>%
) for chaining operationsdplyr
and tidyr
: select
,
filter
, group_by
,
nest
/unnest
,
spread
/gather
, etcpurrr
: mapping functions with
map()
We’ll go through some useful techniques for getting and manipulating groups of data.
Load the tidyquant
package to get started.
A very basic example is retrieving the stock prices for multiple stocks. There are three primary ways to do this:
## # A tibble: 756 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AAPL 2016-01-04 25.7 26.3 25.5 26.3 270597600 23.9
## 2 AAPL 2016-01-05 26.4 26.5 25.6 25.7 223164000 23.3
## 3 AAPL 2016-01-06 25.1 25.6 25.0 25.2 273829600 22.8
## 4 AAPL 2016-01-07 24.7 25.0 24.1 24.1 324377600 21.9
## 5 AAPL 2016-01-08 24.6 24.8 24.2 24.2 283192000 22.0
## 6 AAPL 2016-01-11 24.7 24.8 24.3 24.6 198957600 22.3
## 7 AAPL 2016-01-12 25.1 25.2 24.7 25.0 196616800 22.7
## 8 AAPL 2016-01-13 25.1 25.3 24.3 24.3 249758400 22.1
## 9 AAPL 2016-01-14 24.5 25.1 23.9 24.9 252680400 22.6
## 10 AAPL 2016-01-15 24.0 24.4 23.8 24.3 319335600 22.0
## # ℹ 746 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
The output is a single level tibble with all or the stock prices in
one tibble. The auto-generated column name is “symbol”, which can be
preemptively renamed by giving the vector a name
(e.g. stocks <- c("AAPL", "GOOG", "META")
) and then
piping to tq_get
.
First, get a stock list in data frame format either by making the
tibble or retrieving from tq_index
/
tq_exchange
. The stock symbols must be in the first
column.
stock_list <- tibble(stocks = c("AAPL", "JPM", "CVX"),
industry = c("Technology", "Financial", "Energy"))
stock_list
## # A tibble: 3 × 2
## stocks industry
## <chr> <chr>
## 1 AAPL Technology
## 2 JPM Financial
## 3 CVX Energy
TQ03-scaling-and-modeling-with-tidyquant.R
Second, send the stock list to tq_get
. Notice how the
symbol and industry columns are automatically expanded the length of the
stock prices.
## # A tibble: 756 × 9
## stocks industry date open high low close volume adjusted
## <chr> <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AAPL Technology 2016-01-04 25.7 26.3 25.5 26.3 270597600 23.9
## 2 AAPL Technology 2016-01-05 26.4 26.5 25.6 25.7 223164000 23.3
## 3 AAPL Technology 2016-01-06 25.1 25.6 25.0 25.2 273829600 22.8
## 4 AAPL Technology 2016-01-07 24.7 25.0 24.1 24.1 324377600 21.9
## 5 AAPL Technology 2016-01-08 24.6 24.8 24.2 24.2 283192000 22.0
## 6 AAPL Technology 2016-01-11 24.7 24.8 24.3 24.6 198957600 22.3
## 7 AAPL Technology 2016-01-12 25.1 25.2 24.7 25.0 196616800 22.7
## 8 AAPL Technology 2016-01-13 25.1 25.3 24.3 24.3 249758400 22.1
## 9 AAPL Technology 2016-01-14 24.5 25.1 23.9 24.9 252680400 22.6
## 10 AAPL Technology 2016-01-15 24.0 24.4 23.8 24.3 319335600 22.0
## # ℹ 746 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
Get an index…
## # A tibble: 31 × 8
## symbol company identifier sedol weight sector shares_held local_currency
## <chr> <chr> <chr> <chr> <dbl> <chr> <dbl> <chr>
## 1 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 2 GS GOLDMAN SAC… 38141G104 2407… 0.0806 - 5532262 USD
## 3 MSFT MICROSOFT C… 594918104 2588… 0.0656 - 5532262 USD
## 4 HD HOME DEPOT … 437076102 2434… 0.0583 - 5532262 USD
## 5 CAT CATERPILLAR… 149123101 2180… 0.0559 - 5532262 USD
## 6 AMGN AMGEN INC 031162100 2023… 0.0526 - 5532262 USD
## 7 MCD MCDONALD S … 580135101 2550… 0.0457 - 5532262 USD
## 8 V VISA INC CL… 92826C839 B2PZ… 0.0436 - 5532262 USD
## 9 AXP AMERICAN EX… 025816109 2026… 0.0413 - 5532262 USD
## 10 CRM SALESFORCE … 79466L302 2310… 0.0408 - 5532262 USD
## # ℹ 21 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
…or, get an exchange.
TQ03-scaling-and-modeling-with-tidyquant.R
Send the index or exchange to tq_get
. Important
Note: This can take several minutes depending on the size of the index
or exchange, which is why only the first three stocks are evaluated in
the vignette.
## # A tibble: 8,052 × 15
## symbol company identifier sedol weight sector shares_held local_currency
## <chr> <chr> <chr> <chr> <dbl> <chr> <dbl> <chr>
## 1 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 2 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 3 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 4 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 5 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 6 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 7 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 8 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 9 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## 10 UNH UNITEDHEALT… 91324P102 2917… 0.0940 - 5532262 USD
## # ℹ 8,042 more rows
## # ℹ 7 more variables: date <date>, open <dbl>, high <dbl>, low <dbl>,
## # close <dbl>, volume <dbl>, adjusted <dbl>
TQ03-scaling-and-modeling-with-tidyquant.R
You can use any applicable “getter” to get data for every stock in an index or an exchange! This includes: “stock.prices”, “key.ratios”, “key.stats”, and more.
Once you get the data, you typically want to do something with it.
You can easily do this at scale. Let’s get the yearly returns for
multiple stocks using tq_transmute
. First, get the prices.
We’ll use the FANG
data set, but you typically will use
tq_get
to retrieve data in “tibble” format.
## # A tibble: 4,032 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 META 2013-01-02 27.4 28.2 27.4 28 69846400 28
## 2 META 2013-01-03 27.9 28.5 27.6 27.8 63140600 27.8
## 3 META 2013-01-04 28.0 28.9 27.8 28.8 72715400 28.8
## 4 META 2013-01-07 28.7 29.8 28.6 29.4 83781800 29.4
## 5 META 2013-01-08 29.5 29.6 28.9 29.1 45871300 29.1
## 6 META 2013-01-09 29.7 30.6 29.5 30.6 104787700 30.6
## 7 META 2013-01-10 30.6 31.5 30.3 31.3 95316400 31.3
## 8 META 2013-01-11 31.3 32.0 31.1 31.7 89598000 31.7
## 9 META 2013-01-14 32.1 32.2 30.6 31.0 98892800 31.0
## 10 META 2013-01-15 30.6 31.7 29.9 30.1 173242600 30.1
## # ℹ 4,022 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
Second, use group_by
to group by stock symbol. Third,
apply the mutation. We can do this in one easy workflow. The
periodReturn
function is applied to each group of stock
prices, and a new data frame was returned with the annual returns in the
correct periodicity.
FANG_returns_yearly <- FANG %>%
group_by(symbol) %>%
tq_transmute(select = adjusted,
mutate_fun = periodReturn,
period = "yearly",
col_rename = "yearly.returns")
TQ03-scaling-and-modeling-with-tidyquant.R
Last, we can visualize the returns.
FANG_returns_yearly %>%
ggplot(aes(x = year(date), y = yearly.returns, fill = symbol)) +
geom_bar(position = "dodge", stat = "identity") +
labs(title = "FANG: Annual Returns",
subtitle = "Mutating at scale is quick and easy!",
y = "Returns", x = "", color = "") +
scale_y_continuous(labels = scales::percent) +
coord_flip() +
theme_tq() +
scale_fill_tq()
TQ03-scaling-and-modeling-with-tidyquant.R
Eventually you will want to begin modeling (or more generally
applying functions) at scale! One of the best features
of the tidyverse
is the ability to map functions to nested
tibbles using purrr
. From the Many Models chapter of “R for Data Science”, we can apply the
same modeling workflow to financial analysis. Using a two step
workflow:
Let’s go through an example to illustrate.
In this example, we’ll use a simple linear model to identify the trend in annual returns to determine if the stock returns are decreasing or increasing over time.
First, let’s collect stock data with tq_get()
## # A tibble: 2,518 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AAPL 2007-01-03 3.08 3.09 2.92 2.99 1238319600 2.53
## 2 AAPL 2007-01-04 3.00 3.07 2.99 3.06 847260400 2.58
## 3 AAPL 2007-01-05 3.06 3.08 3.01 3.04 834741600 2.57
## 4 AAPL 2007-01-08 3.07 3.09 3.05 3.05 797106800 2.58
## 5 AAPL 2007-01-09 3.09 3.32 3.04 3.31 3349298400 2.79
## 6 AAPL 2007-01-10 3.38 3.49 3.34 3.46 2952880000 2.93
## 7 AAPL 2007-01-11 3.43 3.46 3.40 3.42 1440252800 2.89
## 8 AAPL 2007-01-12 3.38 3.39 3.33 3.38 1312690400 2.85
## 9 AAPL 2007-01-16 3.42 3.47 3.41 3.47 1244076400 2.93
## 10 AAPL 2007-01-17 3.48 3.49 3.39 3.39 1646260000 2.86
## # ℹ 2,508 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
Next, come up with a function to help us collect annual log returns.
The function below mutates the stock prices to period returns using
tq_transmute()
. We add the type = "log"
and
period = "monthly"
arguments to ensure we retrieve a tibble
of monthly log returns. Last, we take the mean of the monthly returns to
get MMLR.
get_annual_returns <- function(stock.returns) {
stock.returns %>%
tq_transmute(select = adjusted,
mutate_fun = periodReturn,
type = "log",
period = "yearly")
}
TQ03-scaling-and-modeling-with-tidyquant.R
Let’s test get_annual_returns
out. We now have the
annual log returns over the past ten years.
## # A tibble: 10 × 2
## date yearly.returns
## <date> <dbl>
## 1 2007-12-31 0.860
## 2 2008-12-31 -0.842
## 3 2009-12-31 0.904
## 4 2010-12-31 0.426
## 5 2011-12-30 0.228
## 6 2012-12-31 0.282
## 7 2013-12-31 0.0776
## 8 2014-12-31 0.341
## 9 2015-12-31 -0.0306
## 10 2016-12-30 0.118
TQ03-scaling-and-modeling-with-tidyquant.R
Let’s visualize to identify trends. We can see from the linear trend line that AAPL’s stock returns are declining.
AAPL_annual_log_returns %>%
ggplot(aes(x = year(date), y = yearly.returns)) +
geom_hline(yintercept = 0, color = palette_light()[[1]]) +
geom_point(size = 2, color = palette_light()[[3]]) +
geom_line(linewidth = 1, color = palette_light()[[3]]) +
geom_smooth(method = "lm", se = FALSE) +
labs(title = "AAPL: Visualizing Trends in Annual Returns",
x = "", y = "Annual Returns", color = "") +
theme_tq()
TQ03-scaling-and-modeling-with-tidyquant.R
Now, we can get the linear model using the lm()
function. However, there is one problem: the output is not “tidy”.
##
## Call:
## lm(formula = yearly.returns ~ year(date), data = AAPL_annual_log_returns)
##
## Coefficients:
## (Intercept) year(date)
## 58.86282 -0.02915
TQ03-scaling-and-modeling-with-tidyquant.R
We can utilize the broom
package to get “tidy” data from
the model. There’s three primary functions:
augment
: adds columns to the original data such as
predictions, residuals and cluster assignmentsglance
: provides a one-row summary of model-level
statisticstidy
: summarizes a model’s statistical findings such as
coefficients of a regressionWe’ll use tidy
to retrieve the model coefficients.
## # A tibble: 2 × 5
## term estimate std.error statistic p.value
## <chr> <dbl> <dbl> <dbl> <dbl>
## 1 (Intercept) 58.9 113. 0.520 0.617
## 2 year(date) -0.0291 0.0562 -0.518 0.618
TQ03-scaling-and-modeling-with-tidyquant.R
Adding to our workflow, we have the following:
get_model <- function(stock_data) {
annual_returns <- get_annual_returns(stock_data)
mod <- lm(yearly.returns ~ year(date), data = annual_returns)
tidy(mod)
}
TQ03-scaling-and-modeling-with-tidyquant.R
Testing it out on a single stock. We can see that the “term” that contains the direction of the trend (the slope) is “year(date)”. The interpretation is that as year increases one unit, the annual returns decrease by 3%.
## # A tibble: 2 × 5
## term estimate std.error statistic p.value
## <chr> <dbl> <dbl> <dbl> <dbl>
## 1 (Intercept) 58.9 113. 0.520 0.617
## 2 year(date) -0.0291 0.0562 -0.518 0.618
TQ03-scaling-and-modeling-with-tidyquant.R
Now that we have identified the trend direction, it looks like we are ready to scale.
Once the analysis for one stock is done scale to many stocks is
simple. For brevity, we’ll randomly sample ten stocks from the
S&P500 with a call to dplyr::sample_n()
.
## # A tibble: 5 × 8
## symbol company identifier sedol weight sector shares_held local_currency
## <chr> <chr> <chr> <chr> <dbl> <chr> <dbl> <chr>
## 1 MHK MOHAWK INDU… 608190104 2598… 1.68e-4 - 612264 USD
## 2 TGT TARGET CORP 87612E106 2259… 1.51e-3 - 5500695 USD
## 3 WST WEST PHARMA… 955306105 2950… 4.95e-4 - 881367 USD
## 4 BAX BAXTER INTE… 071813109 2085… 4.08e-4 - 6041921 USD
## 5 NWSA NEWS CORP … 65249B109 BBGV… 2.26e-4 - 4549095 USD
TQ03-scaling-and-modeling-with-tidyquant.R
We can now apply our analysis function to the stocks using
dplyr::mutate()
and purrr::map()
. The
mutate()
function adds a column to our tibble, and the
map()
function maps our custom get_model
function to our tibble of stocks using the symbol
column.
The tidyr::unnest()
function unrolls the nested data frame
so all of the model statistics are accessible in the top data frame
level. The filter
, arrange
and
select
steps just manipulate the data frame to isolate and
arrange the data for our viewing.
stocks_model_stats <- stocks_tbl %>%
select(symbol, company) %>%
tq_get(from = "2007-01-01", to = "2016-12-31") %>%
# Nest
group_by(symbol, company) %>%
nest() %>%
# Apply the get_model() function to the new "nested" data column
mutate(model = map(data, get_model)) %>%
# Unnest and collect slope
unnest(model) %>%
filter(term == "year(date)") %>%
arrange(desc(estimate)) %>%
select(-term)
stocks_model_stats
## # A tibble: 5 × 7
## # Groups: symbol, company [5]
## symbol company data estimate std.error statistic p.value
## <chr> <chr> <list> <dbl> <dbl> <dbl> <dbl>
## 1 WST WEST PHARMACEUTICAL SERV… <tibble> 0.0524 0.0216 2.43 0.0412
## 2 MHK MOHAWK INDUSTRIES INC <tibble> 0.0420 0.0294 1.43 0.190
## 3 TGT TARGET CORP <tibble> 0.0178 0.0240 0.741 0.480
## 4 BAX BAXTER INTERNATIONAL INC <tibble> 0.00242 0.0160 0.151 0.884
## 5 NWSA NEWS CORP CLASS A <tibble> -0.0819 0.0480 -1.71 0.230
TQ03-scaling-and-modeling-with-tidyquant.R
We’re done! We now have the coefficient of the linear regression that
tracks the direction of the trend line. We can easily extend this type
of analysis to larger lists or stock indexes. For example, the entire
S&P500 could be analyzed removing the sample_n()
following the call to tq_index("SP500")
.
Eventually you will run into a stock index, stock symbol, FRED data code, etc that cannot be retrieved. Possible reasons are:
This becomes painful when scaling if the functions return errors. So,
the tq_get()
function is designed to handle errors
gracefully. What this means is an NA
value is
returned when an error is generated along with a gentle error
warning.
## [1] NA
TQ03-scaling-and-modeling-with-tidyquant.R
There are pros and cons to this approach that you may not agree with, but I believe helps in the long run. Just be aware of what happens:
Pros: Long running scripts are not interrupted because of one error
Cons: Errors can be inadvertently handled or flow downstream if the user does not read the warnings
Let’s see an example when using tq_get()
to get the
stock prices for a long list of stocks with one BAD APPLE
.
The argument complete_cases
comes in handy. The default is
TRUE
, which removes “bad apples” so future analysis have
complete cases to compute on. Note that a gentle warning stating that an
error occurred and was dealt with by removing the rows from the
results.
## Warning: There was 1 warning in `dplyr::mutate()`.
## ℹ In argument: `data.. = purrr::map(...)`.
## Caused by warning:
## ! x = 'BAD APPLE', get = 'stock.prices': Error in getSymbols.yahoo(Symbols = "BAD APPLE", env = <environment>, : Unable to import "BAD APPLE".
## URL rejected: Malformed input to a URL function
## Removing BAD APPLE.
## # A tibble: 5,368 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AAPL 2014-01-02 19.8 19.9 19.7 19.8 234684800 17.3
## 2 AAPL 2014-01-03 19.7 19.8 19.3 19.3 392467600 16.9
## 3 AAPL 2014-01-06 19.2 19.5 19.1 19.4 412610800 17.0
## 4 AAPL 2014-01-07 19.4 19.5 19.2 19.3 317209200 16.8
## 5 AAPL 2014-01-08 19.2 19.5 19.2 19.4 258529600 17.0
## 6 AAPL 2014-01-09 19.5 19.5 19.1 19.2 279148800 16.7
## 7 AAPL 2014-01-10 19.3 19.3 19.0 19.0 304976000 16.6
## 8 AAPL 2014-01-13 18.9 19.4 18.9 19.1 378492800 16.7
## 9 AAPL 2014-01-14 19.2 19.5 19.2 19.5 332561600 17.0
## 10 AAPL 2014-01-15 19.8 20.0 19.7 19.9 391638800 17.4
## # ℹ 5,358 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
Now switching complete_cases = FALSE
will retain any
errors as NA
values in a nested data frame. Notice that the
error message and output change. The error message now states that the
NA
values exist in the output and the return is a “nested”
data structure.
## Warning: There was 1 warning in `dplyr::mutate()`.
## ℹ In argument: `data.. = purrr::map(...)`.
## Caused by warning:
## ! x = 'BAD APPLE', get = 'stock.prices': Error in getSymbols.yahoo(Symbols = "BAD APPLE", env = <environment>, : Unable to import "BAD APPLE".
## URL rejected: Malformed input to a URL function
## # A tibble: 5,369 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AAPL 2014-01-02 19.8 19.9 19.7 19.8 234684800 17.3
## 2 AAPL 2014-01-03 19.7 19.8 19.3 19.3 392467600 16.9
## 3 AAPL 2014-01-06 19.2 19.5 19.1 19.4 412610800 17.0
## 4 AAPL 2014-01-07 19.4 19.5 19.2 19.3 317209200 16.8
## 5 AAPL 2014-01-08 19.2 19.5 19.2 19.4 258529600 17.0
## 6 AAPL 2014-01-09 19.5 19.5 19.1 19.2 279148800 16.7
## 7 AAPL 2014-01-10 19.3 19.3 19.0 19.0 304976000 16.6
## 8 AAPL 2014-01-13 18.9 19.4 18.9 19.1 378492800 16.7
## 9 AAPL 2014-01-14 19.2 19.5 19.2 19.5 332561600 17.0
## 10 AAPL 2014-01-15 19.8 20.0 19.7 19.9 391638800 17.4
## # ℹ 5,359 more rows
TQ03-scaling-and-modeling-with-tidyquant.R
In both cases, the prudent user will review the warnings to determine
what happened and whether or not this is acceptable. In the
complete_cases = FALSE
example, if the user attempts to
perform downstream computations at scale, the computations will likely
fail grinding the analysis to a halt. But, the advantage is that the
user will more easily be able to filter to the problem root to determine
what happened and decide whether this is acceptable or not.