Pointers, Tips, Code, and More
(Kindly note that this auxiliary website is accessible from my University website experts.mcmaster.ca/display/racinej and is not intended as a substitute)
Overview
The information below contains pointers, links, code, and other information that some may find useful. It is restricted mostly to books I have written (Racine (2019a), Racine (2019b), Li and Racine (2007)) or co-authored and software I have written or co-authored in R (Hayfield and Racine (2008), Racine and Nie (2023), Nie and Racine (2012), etc.).
Books
Racine, J.S. (2019), An Introduction to the Advanced Theory and Practice of Nonparametric Econometrics (A Replicable Approach Using R), Cambridge University Press, ISBN 9781108483407, 408 pages.
Note that lecture slides, assignments, exams, and a solutions manual are available upon request to instructors who adopt this book (slides in \(\rm\LaTeX\) Beamer format). See the website link below for further details (see the Companion Website in the link below).
You can order the book directly from Cambridge University Press (Link: An Introduction to the Advanced Theory and Practice of Nonparametric Econometrics) or from your favourite online retailer when available.
Racine, J.S. (2019), Reproducible Econometrics Using R, Oxford University Press, ISBN: 9780190900663, 293 pages.
Here is the Errata (pdf). Note that lecture slides, assignments, exams, and a solutions manual are available upon request to instructors who adopt this book (slides in \(\rm\LaTeX\) Beamer format). See the website link below for further details (see the Companion Website in the link below).
You can order the book directly from Oxford University Press (Link: Reproducible Econometrics Using R) or from your favourite online retailer.
Li, Q. and J.S. Racine (2007), Nonparametric Econometrics: Theory and Practice, Princeton University Press, ISBN: 9780691121611, 746 Pages.
You can order the book directly from Princeton University Press (Link: Nonparametric Econometrics: Theory and Practice) or from your favourite online retailer.
Here is the table of contents (pdf), Chapter 1 (pdf), the Errata (pdf), the solution manual containing code and answers to odd numbered questions (pdf), and R code for answers to all applied questions (zip). A solution manual containing code and answers to all questions (odd and even) is available to instructors upon request. To receive a copy kindly email me your course syllabus along with your surface mailing address. A hard copy will then be sent via surface mail.
Chinese Translation:
Li, Q. and J.S. Racine, Nonparametric Econometrics: Theory and Practice, Translated by Ye Zhong, Wu Xianbgo et al., Peking University Press (2015), ISBN: 9787301249673.
Monographs
Racine, J.S. (2008), Nonparametric Econometrics: A Primer, Foundations and Trends in Econometrics: Vol. 3: No 1, pp 1-88. (Link: https://dx.doi.org/10.1561/0800000009).
Here is the R code to replicate examples in this primer (zip).
Russian Translation:
An edited version of this monograph is reprinted in Russian and appears as Racine, J.S. (2008) “Nonparametric Econometrics: A Primer”, Quantile, Number 4, pp 7-56.
Edited Volumes
Oxford Handbook of Semiparametric and Nonparametric Econometric Methods, ISBN 978–0–19–985794–4, Edited By Jeffrey S. Racine, Liangjun Su, and Aman Ullah, Published: 2014.
Advances In Econometrics: Nonparametric Econometric Methods, Volume 25, ISBN: 978-1-84950-623-6, Edited by: Qi Li, Jeffrey S. Racine, Published: 2009.
The R np and npRmpi Packages
Consult the np FAQ (pdf) for responses to commonly asked questions and the user manual (pdf) for functions, descriptions, and examples.
The R (www.r-project.org) np and npRmpi packages (current version 0.60-8) implement a variety of nonparametric and semiparametric kernel-based methods in R, an open source platform for statistical computing and graphics. Methods include kernel regression, kernel density estimation, kernel conditional density estimation, and a range of inferential procedures. See the links to the vignettes below for an overview of both packages (I would advise starting with the np vignette).
The np package is the standard package you would use under R, while the npRmpi package is a version that uses the message passing interface for parallel computing. The npRmpi package is designed for executing batch programs on compute clusters and multi-core computing environments to reduce the run time for computationally intensive jobs. See the example files in the demo directory of the npRmpi package for illustrative npRmpi code, and see the examples in the help files and the link for replicating examples for the primer above for code to generate a range of illustrative examples.
Here is a direct link to the np package on the Comprehensive R Archive Network (CRAN), a direct link to the npRmpi package on CRAN, a direct link to the CHANGELOG file on CRAN (documents differences between all versions), an npRmpi test file `test.R’ (text), the npRmpi .Rprofile file (text), install instructions for npRmpi under Windows (text), and instructions for compiling the npRmpi binary from scratch under Windows (text), and instructions for compiling the npRmpi source from scratch for Mac OS X Mountain Lion (text). See the npRmpi GitHub repository (link below) for a recent npRmpi MS Windows binary (available as a binary zip file from the GitHub Downloads menu) and a recent npRmpi Mac OS X binary (available as a binary tgz file from the GitHub Downloads menu).
See the October 2007 Rnews article (pdf) that describes the np package, the np vignette (pdf) for an overview of the np package, the npRmpi vignette (pdf) for an overview of the installation and use of the npRmpi package, and the entropy-based inference vignette for an overview of computing entropy measures (pdf) (R code).
See also the review of the np package that appeared in 2008 in the Journal of Applied Econometrics (link to article in the Wiley Online Library) and the review of the npRmpi package that appeared in 2011 in the Journal of Applied Econometrics (link to article in the Wiley Online Library).
These packages are hosted on GitHub (link).
The R crs Package
Consult the crs FAQ (pdf) for responses to commonly asked questions and the user manual (pdf) for functions, descriptions, and examples.
The R (www.r-project.org) crs package (current version 0.15-31) implements multivariate regression splines (and quantile regression splines as of version 0.15-8) with both continuous and categorical predictors in R, an open source platform for statistical computing and graphics. See the links to the vignettes below for an overview of the package.
Here is a direct link to the crs package on the Comprehensive R Archive Network (CRAN), a direct link to the CHANGELOG file on CRAN (documents differences between all versions).
See the R Journal article (pdf) that describes the crs package, the crs vignette (pdf) for an overview of the crs package and the spline primer vignette for an overview of regression splines (pdf).
This package is hosted on GitHub (link).
The R ma Package
Consult the user manual (pdf) for functions, descriptions, and examples, and the vignette (pdf) for an overview.
The R (www.r-project.org) ma package (current version 1.0-8) implements model averaging using a variety of multivariate bases and averaging criteria.
This package is hosted on GitHub (link).
The R hr Package
Consult the user manual (pdf) for functions, descriptions, and examples.
The R (www.r-project.org) ma package (current version 1.0-1) implements the Hansen-Racine bootstrap model average unit root test.
This package is hosted on GitHub (link).
Stata Implementations of Mixed Data Kernel Estimation (Racine & Li (2004) Journal of Econometrics, Li & Racine (2004) Statistica Sinica)
As of Stata Version 15, Stata users now have access to our mixed-datatype kernel regression methods. See https://www.stata.com/manuals15/rnpregress.pdf for details. The Stata 15 function npregress is the counterpart to the function npreg in the R package np, albeit with different defaults and options.
Gallery of Code and Applications for the np, npRmpi, and crs R Packages
The following link (link to gallery) will take you to a gallery where you can find some commented examples of working code for a range of estimators contained in the np, npRmpi, and crs packages outlined below. Feel free to email me with suggestions. I welcome code/examples that can be showcased and shared with other users, so please feel free to send me code that you would like to share and I will host it in the gallery along with your contact information.