Package {OpenSourceAP.DownloadR}

Title:	Download Open Source Asset Pricing (OpenAP) Data Directly
Version:	0.2.0
Description:	Convenient download functions enabling access Open Source Asset Pricing (OpenAP) data. This package enables users to download predictor portfolio returns (over 200 cross-sectional predictors with multiple portfolio construction methods) and firm characteristics (over 200 characteristics replicated from the academic asset pricing literature). Center for Research in Security Prices (CRSP)-based variables such as Price, Size, and Short-term Reversal can be downloaded with a Wharton Research Data Services (WRDS, https://wrds-www.wharton.upenn.edu/) subscription. For a full list of what is available, see https://www.openassetpricing.com/. Data are read from the OpenAP Hugging Face dataset https://huggingface.co/datasets/OpenSourceAP/CrossSection, which holds data releases from 2020 onward.
License:	MIT + file LICENSE
Encoding:	UTF-8
RoxygenNote:	7.3.3
Suggests:	knitr, rmarkdown, testthat (≥ 3.0.0)
Config/testthat/edition:	3
Depends:	R (≥ 4.1.0)
Imports:	httr, curl, jsonlite, nanoparquet, dplyr, R6, DBI, RPostgres, getPass
NeedsCompilation:	no
Packaged:	2026-10-26 11:53:15 UTC; Tom
Author:	Tom Zimmermann [aut, cre], Justus Boeker [ctb]
Maintainer:	Tom Zimmermann <tom.zimmermann@uni-koeln.de>
Repository:	CRAN
Date/Publication:	2026-10-26 13:10:15 UTC

OpenAP Download

Description

Provides access to data from the Open Source Asset Pricing (OpenAP) project directly in R. The package enables users to access two primary types of data:

1. Predictor Portfolio Returns:

   • Access 212 cross-sectional predictors.

   • Download portfolio returns using various portfolio construction methods, including original paper methods, deciles, quintiles, equal-weighted, value-weighted, price filters and more.

2. Firm Characteristics:

   • Access 209 firm characteristics from OpenAP (+ 3 additional characteristics from CRSP (Price, Size, Short-term Reversal)).

Data are read from the OpenAP Hugging Face dataset (https://huggingface.co/datasets/OpenSourceAP/CrossSection), where each data release is a branch. Releases 2020.05 through 2026.10 are available; 2026.10 is the default.

Learn more about OpenAP: Data website | GitHub code | Publication: doi:10.1561/104.00000112

Public fields

Methods

Public methods

• OpenAP$new()

• OpenAP$list_port()

• OpenAP$get_url()

• OpenAP$get_urls()

• OpenAP$dl_port()

• OpenAP$dl_signal_crsp3()

• OpenAP$merge_crsp_with_signals()

• OpenAP$dl_signal()

• OpenAP$signal_urls()

• OpenAP$dl_all_signals()

• OpenAP$dl_signal_doc()

• OpenAP$clone()

Method new()

Initializes the OpenAP class instance with data for the specified release (or per default with the 2026.10 release). Loads the release's file list and signal documentation.

Usage

OpenAP$new(release_year = NULL, mock = FALSE)

Arguments

Examples

# Offline example
obj <- OpenAP$new(mock = TRUE)

# Real initialization (requires internet connection)
\dontrun{
  openap <- OpenAP$new()
  openap_24 <- OpenAP$new(release_year = "2024_10")
}

Method list_port()

Returns a list of available portfolio types for the OpenAP dataset, depending on the specified release year.

Usage

OpenAP$list_port()

Examples

# List available portfolios  (offline example)
openap <- OpenAP$new(mock = TRUE)
openap$list_port()

# Real usage (internet connection required)
\dontrun{
  openap <- OpenAP$new()
  openap$list_port()
}

Method get_url()

Retrieves the URL for a specific dataset based on its name.

Usage

OpenAP$get_url(data_name)

Arguments

Method get_urls()

Retrieves the URLs for a dataset. Unsigned firm characteristics are two files in the 2020 releases; every other dataset is one file.

Usage

OpenAP$get_urls(data_name)

Arguments

Method dl_port()

Downloads portfolio data for a specified data set and optionally filters by predictor.

Usage

OpenAP$dl_port(data_name, predictor = NULL)

Arguments

Examples

# Download entire portfolio file (offline example)
openap <- OpenAP$new(mock = TRUE)
df <- openap$dl_port("op")
# Download with specific predictor(s) only
df <- openap$dl_port("op", predictor = "Accruals")

# Real usage (internet connection required)
\dontrun{
  openap <- OpenAP$new()
  df <- openap$dl_port("op", predictor = "Accruals")
  df <- openap$dl_port("op", predictor = c("Accruals", "Mom12m"))
}

Method dl_signal_crsp3()

Is being used in "dl_signal()" to download CRSP signals. Requires WRDS credentials. Reads the CRSP CIZ monthly stock file (crsp.msf_v2).

Usage

OpenAP$dl_signal_crsp3(
  requested_crsp_signals = c("Price", "Size", "STreversal"),
  signed = TRUE
)

Arguments

Method merge_crsp_with_signals()

Merges downloaded OpenAP firm-level signals with CRSP-based signals (e.g., Size, Price, STreversal).

Usage

OpenAP$merge_crsp_with_signals(signals, crsp_data)

Arguments

Method dl_signal()

Downloads specific firm characteristics. Only the requested columns of the release's firm characteristics file are downloaded.

Usage

OpenAP$dl_signal(predictor = NULL, signed = FALSE)

Arguments

Returns

A data frame containing the signal data.

Examples

# Download firm characteristics (offline example)
openap <- OpenAP$new(mock = TRUE)

# Single signal (OpenAP or WRDS)
df <- openap$dl_signal("BM")

# Multiple signals
df <- openap$dl_signal(c("BM", "STreversal"))

# Real usage (requires internet connection)
\dontrun{
  openap <- OpenAP$new()
  df <- openap$dl_signal("BM")
  df <- openap$dl_signal(c("BM", "STreversal"))
  df <- openap$dl_signal("BM", signed = TRUE)
}

Method signal_urls()

Retrieves the URLs of the firm characteristics file(s) of the release.

Usage

OpenAP$signal_urls(signed)

Arguments

Method dl_all_signals()

Downloads all firm level characteristics from the release folder.

Usage

OpenAP$dl_all_signals(signed = TRUE)

Arguments

Returns

A data frame containing all firm level characteristics.

Examples

# Download all signals (offline example)
openap <- OpenAP$new(mock = TRUE)
signals_data <- openap$dl_all_signals()

# Real usage (requires internet connection)
\dontrun{
  openap <- OpenAP$new()
  signals_data <- openap$dl_all_signals()
}

Method dl_signal_doc()

Downloads the signal documentation CSV for the release.

Usage

OpenAP$dl_signal_doc()

Returns

A data frame containing the signal documentation.

Examples

# Load signal documentation (offline example)
openap <- OpenAP$new(mock = TRUE)
signal_doc <- openap$dl_signal_doc()
# Real usage (requires internet connection)
\dontrun{
  openap <- OpenAP$new()
  signal_doc <- openap$dl_signal_doc()
}

Method clone()

The objects of this class are cloneable with this method.

Usage

OpenAP$clone(deep = FALSE)

Arguments

Examples

## ------------------------------------------------
## Method `OpenAP$new`
## ------------------------------------------------

# Offline example
obj <- OpenAP$new(mock = TRUE)

# Real initialization (requires internet connection)
## Not run: 
  openap <- OpenAP$new()
  openap_24 <- OpenAP$new(release_year = "2024_10")

## End(Not run)

## ------------------------------------------------
## Method `OpenAP$list_port`
## ------------------------------------------------

# List available portfolios  (offline example)
openap <- OpenAP$new(mock = TRUE)
openap$list_port()

# Real usage (internet connection required)
## Not run: 
  openap <- OpenAP$new()
  openap$list_port()

## End(Not run)

## ------------------------------------------------
## Method `OpenAP$dl_port`
## ------------------------------------------------

# Download entire portfolio file (offline example)
openap <- OpenAP$new(mock = TRUE)
df <- openap$dl_port("op")
# Download with specific predictor(s) only
df <- openap$dl_port("op", predictor = "Accruals")

# Real usage (internet connection required)
## Not run: 
  openap <- OpenAP$new()
  df <- openap$dl_port("op", predictor = "Accruals")
  df <- openap$dl_port("op", predictor = c("Accruals", "Mom12m"))

## End(Not run)

## ------------------------------------------------
## Method `OpenAP$dl_signal`
## ------------------------------------------------

# Download firm characteristics (offline example)
openap <- OpenAP$new(mock = TRUE)

# Single signal (OpenAP or WRDS)
df <- openap$dl_signal("BM")

# Multiple signals
df <- openap$dl_signal(c("BM", "STreversal"))

# Real usage (requires internet connection)
## Not run: 
  openap <- OpenAP$new()
  df <- openap$dl_signal("BM")
  df <- openap$dl_signal(c("BM", "STreversal"))
  df <- openap$dl_signal("BM", signed = TRUE)

## End(Not run)

## ------------------------------------------------
## Method `OpenAP$dl_all_signals`
## ------------------------------------------------

# Download all signals (offline example)
openap <- OpenAP$new(mock = TRUE)
signals_data <- openap$dl_all_signals()

# Real usage (requires internet connection)
## Not run: 
  openap <- OpenAP$new()
  signals_data <- openap$dl_all_signals()

## End(Not run)

## ------------------------------------------------
## Method `OpenAP$dl_signal_doc`
## ------------------------------------------------

# Load signal documentation (offline example)
openap <- OpenAP$new(mock = TRUE)
signal_doc <- openap$dl_signal_doc()
# Real usage (requires internet connection)
## Not run: 
  openap <- OpenAP$new()
  signal_doc <- openap$dl_signal_doc()

## End(Not run)