capybara

About

tldr; If you have a 2-4GB dataset and you need to estimate a (generalized) linear model with a large number of fixed effects, this package is for you. It works with larger datasets as well and facilites computing clustered standard errors.

‘capybara’ is a fast and small footprint software that provides efficient functions for demeaning variables before conducting a GLM estimation. This technique is particularly useful when estimating linear models with multiple group fixed effects. It is a fork of the excellent Alpaca package created and maintained by Dr. Amrei Stammann. The software can estimate Exponential Family models (e.g., Poisson) and Negative Binomial models.

Traditional QR estimation can be unfeasible due to additional memory requirements. The method, which is based on Halperin (1962) vector projections offers important time and memory savings without compromising numerical stability in the estimation process.

The software heavily borrows from Gaure (2013) and Stammann (2018) works on OLS and GLM estimation with large fixed effects implemented in the ‘lfe’ and ‘alpaca’ packages. The differences are that ‘capybara’ does not use C nor Rcpp code, instead it uses cpp11 and cpp11armadillo.

The summary tables borrow from Stata outputs. I have also provided integrations with ‘broom’ to facilitate the inclusion of statistical tables in Quarto/Jupyter notebooks.

If this software is useful to you, please consider donating on Buy Me A Coffee. All donations will be used to continue improving capybara.

Installation

You can install the development version of capybara like so:

remotes::install_github("pachadotdev/capybara")

Examples

See the documentation in progress: https://pacha.dev/capybara/.

Design choices

Capybara is full of trade-offs. I have used ‘data.table’ to benefit from in-place modifications. The model fitting is done on C++ side. While the code aims to be fast, I prefer to have some bottlenecks instead of low numerical stability. The principle was: “He who gives up code safety for code speed deserves neither.” (Wickham, 2014).

Benchmarks

Median time and memory footprint for the different models in the book An Advanced Guide to Trade Policy Analysis.

Changing the number of cores

Note that you can edit the Makevars file to change the number of cores that capybara uses, here is an example of how it affects the performance

Testing and debugging

Testing

I use testthat (e.g., devtools::test()) to compare the results with base R. These tests are about the correctness of the results.

function r_debug_symbols () {
    # if src/Makevars does not exist, exit
    if [ ! -f src/Makevars ]; then
        echo "File src/Makevars does not exist"
        return 1
    fi

    # if src/Makevars contains a line that says "PKG_CPPFLAGS"
    # but there is no "-UDEBUG -g" on it
    # then add "PKG_CPPFLAGS += -UDEBUG -g" at the end
    if grep -q "PKG_CPPFLAGS" src/Makevars; then
        if ! grep -q "PKG_CPPFLAGS.*-UDEBUG.*-g" src/Makevars; then
            echo "PKG_CPPFLAGS += -UDEBUG -g" >> src/Makevars
        fi
    fi

    # if src/Makevars does not contain a line that reads
    # PKG_CPPFLAGS ...something... -UDEBUG -g ...something...
    # then add PKG_CPPFLAGS = -UDEBUG -g to it
    if ! grep -q "PKG_CPPFLAGS.*-UDEBUG.*-g" src/Makevars; then
        echo "PKG_CPPFLAGS = -UDEBUG -g" >> src/Makevars
    fi
}

function r_valgrind () {
    # if no argument is provided, ask for a file
    if [ -z "$1" ]; then
        read -p "Enter the script to debug: " script
    else
        script=$1
    fi

    # if no output file is provided, use the same filename but ended in txt
    if [ -z "$2" ]; then
        output="${script%.*}.txt"
    else
        output=$2
    fi

    # if the file does not exist, exit
    if [ ! -f "$script" ]; then
        echo "File $script does not exist"
        return 1
    fi

    # if the file does not end in .R/.r, exit
    shopt -s nocasematch
    if [[ "$script" != *.R ]]; then
        echo "File $script does not end in .R or .r"
        return 1
    fi
    shopt -u nocasematch

    # run R in debug mode, but after that we compiled with debug symbols
    # see https://reside-ic.github.io/blog/debugging-memory-errors-with-valgrind-and-gdb/
    # R -d 'valgrind -s --leak-check=full --show-leak-kinds=all --track-origins=yes' -f $script 2>&1 | tee valgrind.txt
    R --vanilla -d 'valgrind -s --track-origins=yes' -f $script 2>&1 | tee $output
}

# create an alias for R
alias r="R"
alias rvalgrind="R --vanilla -d 'valgrind -s --track-origins=yes'"

Code of Conduct

Please note that the capybara project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

Acknowledgements

Thanks a lot to Prof. Yoto Yotov for reviewing the summary functions.