Intro to Tidyquant in R
tidyquant pulls data directly from several different financial and economic sources into R and integrates the core financial packages zoo, xts, quantmod, TTR, and PerformanceAnalytics with the tidyverse syntax.
Load libraries
rm(list=ls())clears all objects from the global environmentgc()frees up memory by cleaning up unused objects; important if you’re working on more memory-intensive analysis
rm(list=ls())
gc()## used (Mb) gc trigger (Mb) limit (Mb) max used (Mb)
## Ncells 531578 28.4 1184037 63.3 NA 669514 35.8
## Vcells 980710 7.5 8388608 64.0 18432 1851810 14.2library(tidyquant)
library(tidyverse)
library(DataExplorer)Pull data
tq_get() pulls web-based financial data from different
sources. Some require an API key.
Yahoo Finance
stock.pricesandstock.prices.japan: open, high, low, close, volume and adjusted stock prices for a stock symboldividends: dividends for a stock symbolsplits: split ratio for a stock symbol
FRED
economic.data: economic data from FRED.
Requires API key and includes crypto and other more specialized and exotic datasets:*
- quandl
- tiingo
- alphavantager
- rblpapi accesses data and calculations from Bloomberg
tq_get_options()## [1] "stock.prices" "stock.prices.japan" "dividends"
## [4] "splits" "economic.data" "quandl"
## [7] "quandl.datatable" "tiingo" "tiingo.iex"
## [10] "tiingo.crypto" "alphavantager" "alphavantage"
## [13] "rblpapi"FRED
Pull economic data from FRED using the series code on the top-right of the chart:
Canada vs. China imports
- Symbol for U.S. Imports of Goods by Customs Basis from Canada: “IMPCA”
- Symbol for U.S. Imports of Goods by Customs Basis from China: “IMPCH”
tq_get(c("IMPCA", "IMPCH"), get="economic.data")## # A tibble: 242 × 3
## symbol date price
## <chr> <date> <dbl>
## 1 IMPCA 2015-01-01 25844.
## 2 IMPCA 2015-02-01 23266.
## 3 IMPCA 2015-03-01 25778.
## 4 IMPCA 2015-04-01 24763.
## 5 IMPCA 2015-05-01 24283.
## 6 IMPCA 2015-06-01 27352.
## 7 IMPCA 2015-07-01 24718.
## 8 IMPCA 2015-08-01 24817.
## 9 IMPCA 2015-09-01 25289.
## 10 IMPCA 2015-10-01 23669.
## # ℹ 232 more rowstq_get(c("IMPCA", "IMPCH"), get="economic.data") %>%
ggplot(aes(x=date, y=price, color=symbol))+
geom_line() +
ggtitle("U.S. Imports of Goods by Customs Basis from China and Canada") +
ylab("Millions of Dollars") +
scale_y_continuous(labels = scales::dollar_format(prefix="$", suffix = "M"))
Stock Prices
open, high, low, and close:the opening, high, low, and closing stock prices that day.[volume:](https://www.investopedia.com/articles/technical/02/010702.asp#:~:text=Volume%20measures%20the%20number%20of,prices%20fall%20on%20increasing%20volume.)the number of trades that day.adjusted stock price:While the closing price simply refers to the cost of shares at the end of the day, the adjusted closing price takes dividends, stock splits, and new stock offerings into account.
Apple
Jan. 1, 1980 is entered as the start date, but Apple went public on Dec. 12, 1980, so this is the earliest date pulled.
tq_get("AAPL", get = "stock.prices",
from = " 1980-01-01")## # A tibble: 11,149 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 AAPL 1980-12-12 0.128 0.129 0.128 0.128 469033600 0.0987
## 2 AAPL 1980-12-15 0.122 0.122 0.122 0.122 175884800 0.0936
## 3 AAPL 1980-12-16 0.113 0.113 0.113 0.113 105728000 0.0867
## 4 AAPL 1980-12-17 0.116 0.116 0.116 0.116 86441600 0.0889
## 5 AAPL 1980-12-18 0.119 0.119 0.119 0.119 73449600 0.0914
## 6 AAPL 1980-12-19 0.126 0.127 0.126 0.126 48630400 0.0970
## 7 AAPL 1980-12-22 0.132 0.133 0.132 0.132 37363200 0.102
## 8 AAPL 1980-12-23 0.138 0.138 0.138 0.138 46950400 0.106
## 9 AAPL 1980-12-24 0.145 0.146 0.145 0.145 48003200 0.112
## 10 AAPL 1980-12-26 0.158 0.159 0.158 0.158 55574400 0.122
## # ℹ 11,139 more rowsMultiple stocks
stocks <- tq_get(c("NVDA", "AMZN", "META", "AAPL"),
get = "stock.prices",
from = "2024-01-01",
to = "2025-03-07")
head(stocks)## # A tibble: 6 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 NVDA 2024-01-02 49.2 49.3 47.6 48.2 411254000 48.2
## 2 NVDA 2024-01-03 47.5 48.2 47.3 47.6 320896000 47.6
## 3 NVDA 2024-01-04 47.8 48.5 47.5 48.0 306535000 48.0
## 4 NVDA 2024-01-05 48.5 49.5 48.3 49.1 415039000 49.1
## 5 NVDA 2024-01-08 49.5 52.3 49.5 52.3 642510000 52.2
## 6 NVDA 2024-01-09 52.4 54.3 51.7 53.1 773100000 53.1For instance, you can see that Nvidia has a much higher volume of trades than the other tech stocks.
stocks %>%
ggplot(aes(x=date, y=volume, color=symbol)) +
geom_line() +
theme_tq() +
#facet_wrap(~symbol, scales="free_y", ncol=2) +
scale_y_continuous(labels = scales::comma_format()) 
Other tq options
tq_exchange_options()## [1] "AMEX" "NASDAQ" "NYSE"tq_fund_source_options()## [1] "SSGA"tq_index_options()## [1] "DOW" "DOWGLOBAL" "SP400" "SP500" "SP600"nyse <- tq_exchange("NYSE")## Getting data...head(nyse)## # A tibble: 6 × 7
## symbol company last.sale.price market.cap country ipo.year industry
## <chr> <chr> <dbl> <dbl> <chr> <int> <chr>
## 1 A "Agilent Technolo… 122. 3.48e10 "Unite… 1999 "Biotec…
## 2 AA "Alcoa Corporatio… 31.0 8.02e 9 "Unite… 2016 "Alumin…
## 3 AACT "Ares Acquisition… 11.1 0 "" 2023 "Blank …
## 4 AAM "AA Mission Acqui… 10.2 0 "" 2024 ""
## 5 AAMI "Acadian Asset Ma… 23.6 8.87e 8 "Unite… 2014 "Invest…
## 6 AAP "Advance Auto Par… 36.6 2.19e 9 "Unite… NA "Auto &…Mutating & Charts
tq_mutate adds columns to the existing dataframe.
tq_transmute works exactly like tq_mutate except it only
returns the newly created columns. This is helpful when changing the
periodicity in the data, such as from daily to quarterly returns, where
the new columns would not have the same number of rows.
More Tutorials:
- Tidyquant’s Quantitative Analysis Package Integrations gives a more in-depth overview of functions and charts.
- PerformanceAnalytics for investors
tq_mutate_fun_options()## $zoo
## [1] "rollapply" "rollapplyr" "rollmax"
## [4] "rollmax.default" "rollmaxr" "rollmean"
## [7] "rollmean.default" "rollmeanr" "rollmedian"
## [10] "rollmedian.default" "rollmedianr" "rollsum"
## [13] "rollsum.default" "rollsumr"
##
## $xts
## [1] "apply.daily" "apply.monthly" "apply.quarterly" "apply.weekly"
## [5] "apply.yearly" "diff.xts" "lag.xts" "period.apply"
## [9] "period.max" "period.min" "period.prod" "period.sum"
## [13] "periodicity" "to.daily" "to.hourly" "to.minutes"
## [17] "to.minutes10" "to.minutes15" "to.minutes3" "to.minutes30"
## [21] "to.minutes5" "to.monthly" "to.period" "to.quarterly"
## [25] "to.weekly" "to.yearly" "to_period"
##
## $quantmod
## [1] "allReturns" "annualReturn" "ClCl" "dailyReturn"
## [5] "Delt" "HiCl" "Lag" "LoCl"
## [9] "LoHi" "monthlyReturn" "Next" "OpCl"
## [13] "OpHi" "OpLo" "OpOp" "periodReturn"
## [17] "quarterlyReturn" "seriesAccel" "seriesDecel" "seriesDecr"
## [21] "seriesHi" "seriesIncr" "seriesLo" "weeklyReturn"
## [25] "yearlyReturn"
##
## $TTR
## [1] "adjRatios" "ADX" "ALMA"
## [4] "aroon" "ATR" "BBands"
## [7] "CCI" "chaikinAD" "chaikinVolatility"
## [10] "CLV" "CMF" "CMO"
## [13] "CTI" "DEMA" "DonchianChannel"
## [16] "DPO" "DVI" "EMA"
## [19] "EMV" "EVWMA" "GMMA"
## [22] "growth" "HMA" "keltnerChannels"
## [25] "KST" "lags" "MACD"
## [28] "MFI" "momentum" "OBV"
## [31] "PBands" "ROC" "rollSFM"
## [34] "RSI" "runCor" "runCov"
## [37] "runMAD" "runMax" "runMean"
## [40] "runMedian" "runMin" "runPercentRank"
## [43] "runSD" "runSum" "runVar"
## [46] "SAR" "SMA" "SMI"
## [49] "SNR" "stoch" "TDI"
## [52] "TRIX" "ultimateOscillator" "VHF"
## [55] "VMA" "volatility" "VWAP"
## [58] "VWMA" "wilderSum" "williamsAD"
## [61] "WMA" "WPR" "ZigZag"
## [64] "ZLEMA"
##
## $PerformanceAnalytics
## [1] "Return.annualized" "Return.annualized.excess"
## [3] "Return.clean" "Return.cumulative"
## [5] "Return.excess" "Return.Geltner"
## [7] "zerofill"Comparing Returns
tq_get(c("NVDA", "META", "AAPL", "MSFT"),
get = "stock.prices",
from = "2024-01-01") %>%
group_by(symbol) %>%
tq_transmute(select = close,
mutate_fun = periodReturn,
period = "quarterly",
type = "arithmetic") %>%
ggplot(aes(x=date, y=quarterly.returns, fill=symbol)) +
geom_col(position = "dodge") +
# facet_wrap(~ symbol, ncol = 2) +
theme_tq() +
scale_fill_tq() +
scale_y_continuous(labels = scales::percent)
Bar Chart
This is a type of bar chart that plots the open, close, high, and low of the daily stock returns and color-codes the bars based on whether the day ended with the stock price up (blue) or down (red).
Source: HowToTrade.com
Tidyquant also has functions for candlestick
charts, geom_candlestick, and for displaying moving
averages, geom_ma() and geom_bbands() for more
complex Bollinger
Bands.
tq_get("NVDA",
get = "stock.prices",
from = "2025-01-01",
to = "2025-03-07") %>%
ggplot(aes(x = date, y = close)) +
geom_barchart(aes(open = open, high = high, low = low, close = close), na.rm = TRUE) +
labs(title = "Nvidia Daily Returns", y = "Closing Price", x = "") +
theme_tq() +
# geom_ma() + # moving average
scale_y_continuous(labels = scales::dollar_format())
Correlation Matrix
The correlation matrix uses the DataExplorer package.
cor <- tq_get(c("META", "NVDA", # Tech
"GLD", "PPLT", # gold and platinum
"AAL", # American airlines
"CVS", # healthcare
"XOM", "DBO"), # Exxon &
get = "stock.prices",
from = "2020-01-01") %>%
group_by(symbol) %>%
tq_transmute(select = close,
mutate_fun = periodReturn,
period = "monthly",
type = "arithmetic") %>%
select(date, symbol, return = monthly.returns) %>%
spread(symbol, return)
cor## # A tibble: 63 × 9
## date AAL CVS DBO GLD META NVDA PPLT XOM
## <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2020-01-31 -0.0773 -0.0855 -0.151 0.0374 -0.0375 -0.0145 -0.0233 -0.124
## 2 2020-02-28 -0.290 -0.127 -0.106 -0.00636 -0.0468 0.142 -0.100 -0.172
## 3 2020-03-31 -0.360 0.00253 -0.245 -0.00222 -0.133 -0.0240 -0.163 -0.262
## 4 2020-04-30 -0.0148 0.0374 -0.0962 0.0726 0.227 0.109 0.0901 0.224
## 5 2020-05-29 -0.126 0.0653 0.173 0.0259 0.0996 0.215 0.0681 -0.0215
## 6 2020-06-30 0.245 -0.00915 0.0754 0.0274 0.00880 0.0701 -0.0113 -0.0165
## 7 2020-07-31 -0.149 -0.0312 0.0486 0.108 0.117 0.118 0.0907 -0.0590
## 8 2020-08-31 0.174 -0.0130 0.0491 -0.00324 0.156 0.260 0.0308 -0.0509
## 9 2020-09-30 -0.0582 -0.0599 -0.0650 -0.0417 -0.107 0.0117 -0.0444 -0.140
## 10 2020-10-30 -0.0822 -0.0396 -0.107 -0.00519 0.00462 -0.0736 -0.0521 -0.0498
## # ℹ 53 more rowsplot_correlation(na.omit(cor, maxcat = 5L))## Warning in dummify(data, maxcat = maxcat): Ignored all discrete features since
## `maxcat` set to 20 categories!