L0 regularization paper (WIP)

paper-l0/.gitignore

@@ -0,0 +1,35 @@
+## Core latex/pdflatex auxiliary files:
+*.aux
+*.lof
+*.log
+*.lot
+*.fls
+*.out
+*.toc
+*.fmt
+*.fot
+*.cb
+*.cb2
+.*.lb
+*.nav
+*.snm
+*.vrb
+
+## Generated if empty string is given at "Please type another file name for output:"
+.pdf
+
+## Bibliography auxiliary files (bibtex/biblatex/biber):
+*.bbl
+*.bcf
+*.blg
+*-blx.aux
+*-blx.bib
+*.run.xml
+
+## Build tool auxiliary files:
+*.fdb_latexmk
+*.synctex
+*.synctex(busy)
+*.synctex.gz
+*.synctex.gz(busy)
+*.pdfsync

paper-l0/bibliography/references.bib

@@ -0,0 +1,219 @@
+% === Core method references ===
+
+@inproceedings{louizos2018,
+  title     = {Learning Sparse Neural Networks through $L_0$ Regularization},
+  author    = {Louizos, Christos and Welling, Max and Kingma, Diederik P.},
+  booktitle = {International Conference on Learning Representations},
+  year      = {2018},
+  url       = {https://openreview.net/forum?id=H1Y8hhg0b}
+}
+
+@article{deville1992,
+  title   = {Calibration Estimators in Survey Sampling},
+  author  = {Deville, Jean-Claude and S{\"a}rndal, Carl-Erik},
+  journal = {Journal of the American Statistical Association},
+  volume  = {87},
+  number  = {418},
+  pages   = {376--382},
+  year    = {1992}
+}
+
+@article{sarndal2007,
+  title   = {The Calibration Approach in Survey Theory and Practice},
+  author  = {S{\"a}rndal, Carl-Erik},
+  journal = {Survey Methodology},
+  volume  = {33},
+  number  = {2},
+  pages   = {99--119},
+  year    = {2007}
+}
+
+@article{deming1940,
+  title   = {On a Least Squares Adjustment of a Sampled Frequency Table When the Expected Marginal Totals are Known},
+  author  = {Deming, W. Edwards and Stephan, Frederick F.},
+  journal = {The Annals of Mathematical Statistics},
+  volume  = {11},
+  number  = {4},
+  pages   = {427--444},
+  year    = {1940}
+}
+
+@article{ireland1968,
+  title   = {Contingency Tables with Given Marginals},
+  author  = {Ireland, C. T. and Kullback, S.},
+  journal = {Biometrika},
+  volume  = {55},
+  number  = {1},
+  pages   = {179--188},
+  year    = {1968}
+}
+
+@inproceedings{kingma2015,
+  title     = {Adam: A Method for Stochastic Optimization},
+  author    = {Kingma, Diederik P. and Ba, Jimmy},
+  booktitle = {International Conference on Learning Representations},
+  year      = {2015},
+  url       = {https://arxiv.org/abs/1412.6980}
+}
+
+% === Spatial microsimulation ===
+
+@article{williamson1998,
+  title   = {The Creation of Census-Based Small Area Microdata Using Statistical Matching Techniques},
+  author  = {Williamson, Paul and Birkin, Mark and Rees, Phil},
+  journal = {Environment and Planning A},
+  volume  = {30},
+  number  = {5},
+  pages   = {785--816},
+  year    = {1998}
+}
+
+@article{huang2001,
+  title   = {A Spatial Microsimulation Model with Student Agents},
+  author  = {Huang, Zhixin and Williamson, Paul},
+  journal = {Computers, Environment and Urban Systems},
+  volume  = {25},
+  number  = {6},
+  pages   = {473--489},
+  year    = {2001}
+}
+
+@article{tanton2011,
+  title   = {Small Area Estimation Using a Reweighting Algorithm},
+  author  = {Tanton, Robert and Vidyattama, Yogi and Nepal, Binod and McNamara, John},
+  journal = {Journal of the Royal Statistical Society: Series A},
+  volume  = {174},
+  number  = {4},
+  pages   = {931--951},
+  year    = {2011}
+}
+
+@book{lovelace2016,
+  title     = {Spatial Microsimulation with {R}},
+  author    = {Lovelace, Robin and Dumont, Morgane},
+  publisher = {Chapman and Hall/CRC},
+  year      = {2016}
+}
+
+@article{harland2012,
+  title   = {Creating Realistic Synthetic Populations at Varying Spatial Scales: A Comparative Critique of Population Synthesis Techniques},
+  author  = {Harland, Kirk and Heppenstall, Alison and Smith, Dianna and Birkin, Mark},
+  journal = {Journal of Artificial Societies and Social Simulation},
+  volume  = {15},
+  number  = {1},
+  pages   = {1},
+  year    = {2012}
+}
+
+@article{anderson2013,
+  title   = {Microsimulation for Local Impact Analysis: The Role of Small Area Estimation},
+  author  = {Anderson, Brent},
+  journal = {International Journal of Microsimulation},
+  volume  = {6},
+  number  = {2},
+  pages   = {30--55},
+  year    = {2013}
+}
+
+% === PolicyEngine and predecessor ===
+
+@techreport{woodruff2024,
+  title       = {Enhancing Survey Microdata with Administrative Records: A Novel Approach to Microsimulation Dataset Construction},
+  author      = {Woodruff, Nikhil and Ghenis, Max},
+  institution = {PolicyEngine},
+  year        = {2024},
+  url         = {https://github.com/PolicyEngine/policyengine-us-data}
+}
+
+% === Data sources ===
+
+@techreport{census2024,
+  title       = {Current Population Survey, 2024 Annual Social and Economic ({ASEC}) Supplement},
+  author      = {{U.S. Census Bureau}},
+  institution = {U.S. Census Bureau},
+  year        = {2024},
+  url         = {https://www2.census.gov/programs-surveys/cps/datasets/2024/march/asec2024_ddl_pub_full.pdf}
+}
+
+@techreport{bryant2023a,
+  title       = {General Description Booklet for the 2015 Public Use Tax File},
+  author      = {Bryant, Victoria},
+  institution = {Statistics of Income Division, Internal Revenue Service},
+  year        = {2023},
+  month       = {February},
+  type        = {Technical Documentation},
+  url         = {https://www.irs.gov/statistics/soi-tax-stats-individual-public-use-microdata-files}
+}
+
+@techreport{cbo2018,
+  title       = {An Overview of {CBO}'s Microsimulation Tax Model},
+  author      = {{Congressional Budget Office}},
+  institution = {Congressional Budget Office},
+  year        = {2018},
+  url         = {https://www.cbo.gov/publication/54096}
+}
+
+@techreport{jct2023,
+  title       = {Overview of {JCT} Revenue Estimating Methods},
+  author      = {{Joint Committee on Taxation}},
+  institution = {Joint Committee on Taxation},
+  number      = {JCX-48-23},
+  year        = {2023},
+  url         = {https://www.jct.gov/publications/2023/jcx-48-23/}
+}
+
+@misc{tpc2024,
+  title  = {Brief Description of the Tax Model},
+  author = {{Tax Policy Center}},
+  year   = {2024},
+  url    = {https://www.taxpolicycenter.org/resources/brief-description-tax-model}
+}
+
+% === Machine learning / imputation ===
+
+@article{meinshausen2006quantile,
+  title   = {Quantile Regression Forests},
+  author  = {Meinshausen, Nicolai},
+  journal = {Journal of Machine Learning Research},
+  volume  = {7},
+  pages   = {983--999},
+  year    = {2006}
+}
+
+@article{pytorch2019,
+  title   = {{PyTorch}: An Imperative Style, High-Performance Deep Learning Library},
+  author  = {Paszke, Adam and Gross, Sam and Massa, Francisco and Lerer, Adam and Bradbury, James and Chanan, Gregory and Killeen, Trevor and Lin, Zeming and Gimelshein, Natalia and Antiga, Luca and others},
+  journal = {Advances in Neural Information Processing Systems},
+  volume  = {32},
+  year    = {2019}
+}
+
+% === Income measurement / data quality ===
+
+@article{burkhauser2012,
+  title   = {Recent Trends in Top Income Shares in the {United States}: Reconciling Estimates from {March CPS} and {IRS} Tax Return Data},
+  author  = {Burkhauser, Richard V. and Feng, Shuaizhang and Jenkins, Stephen P. and Larrimore, Jeff},
+  journal = {Review of Economics and Statistics},
+  volume  = {94},
+  number  = {2},
+  pages   = {371--388},
+  year    = {2012}
+}
+
+@article{rothbaum2021,
+  title   = {Has Income Underreporting Changed Over Time? Evidence from Linked Survey and Administrative Data},
+  author  = {Rothbaum, Jonathan and Bee, Adam},
+  journal = {Journal of Business and Economic Statistics},
+  volume  = {39},
+  number  = {2},
+  pages   = {456--475},
+  year    = {2021}
+}
+
+@article{meyer2021,
+  title   = {The Accuracy of Tax Imputations: Estimating Tax Liabilities and Credits Using Linked Survey and Administrative Data},
+  author  = {Meyer, Bruce D. and Wu, Derek and Finley, Grace and Langetieg, Patrick and Payne, Carla and Plumley, Alan and Yu, Alexa},
+  journal = {NBER Working Paper},
+  number  = {28229},
+  year    = {2021}
+}

paper-l0/ijm.sty

@@ -0,0 +1,69 @@
+% ijm.sty — Style shim for the International Journal of Microsimulation
+% Enforces IJM editorial policy formatting requirements.
+
+\NeedsTeXFormat{LaTeX2e}
+\ProvidesPackage{ijm}[2025/04/01 IJM formatting shim]
+
+% --- Font: Times New Roman equivalent ---
+\RequirePackage{mathptmx}
+
+% --- Spacing: 1.5 throughout ---
+\RequirePackage{setspace}
+\onehalfspacing
+
+% --- Margins: 1.5 inches on all sides ---
+\RequirePackage[margin=1.5in]{geometry}
+
+% --- Section numbering: max 3 levels ---
+\setcounter{secnumdepth}{3}
+\setcounter{tocdepth}{3}
+
+% --- Mathematics ---
+\RequirePackage{amsmath}
+\RequirePackage{amsfonts}
+
+% --- Tables: no borders, Arial 10pt interior ---
+\RequirePackage{booktabs}
+\RequirePackage{array}
+\RequirePackage{helvet}  % provides \sffamily for Arial-like interior text
+
+% Table interior font switch
+\newcommand{\tablefont}{\fontsize{10}{12}\selectfont\sffamily}
+
+% --- Figures ---
+\RequirePackage{graphicx}
+
+% --- Citations: APA style via natbib ---
+\RequirePackage[round]{natbib}
+\bibpunct{(}{)}{;}{a}{,}{,}
+\setcitestyle{authoryear,round}
+
+% --- Hyperlinks ---
+\RequirePackage{hyperref}
+\RequirePackage{xcolor}
+\hypersetup{
+    colorlinks=true,
+    linkcolor=blue,
+    filecolor=magenta,
+    urlcolor=blue,
+    citecolor=blue,
+}
+
+% --- Algorithms ---
+\RequirePackage{algorithm}
+\RequirePackage{algpseudocode}
+
+% --- Footnotes: same font, size 10 ---
+\RequirePackage[hang,flushmargin]{footmisc}
+\renewcommand{\footnotesize}{\fontsize{10}{12}\selectfont}
+
+% --- Float placement ---
+\RequirePackage{float}
+
+% --- Source/note formatting below tables and figures ---
+\newcommand{\tablenote}[1]{%
+  \par\vspace{2pt}%
+  {\fontsize{10}{12}\selectfont\rmfamily #1}%
+}
+
+\endinput

paper-l0/macros.tex

@@ -0,0 +1,30 @@
+% Custom commands and mathematics macros
+
+% --- Acronyms ---
+\newcommand{\policyengine}{\textsc{PolicyEngine}}
+\newcommand{\cps}{\textsc{CPS}}
+\newcommand{\puf}{\textsc{PUF}}
+\newcommand{\acs}{\textsc{ACS}}
+\newcommand{\sipp}{\textsc{SIPP}}
+\newcommand{\scf}{\textsc{SCF}}
+
+% --- L0 / Hard Concrete notation ---
+\newcommand{\lzero}{$L_0$}
+\newcommand{\ltwo}{$L_2$}
+\newcommand{\hc}{Hard Concrete}
+
+% --- Placeholder for empirical results ---
+\newcommand{\tbc}[1][]{%
+  \textbf{[TBC%
+    \def\temparg{#1}%
+    \ifx\temparg\empty\else: #1\fi
+  ]}%
+}
+
+% --- Math shortcuts ---
+\newcommand{\E}{\mathbb{E}}
+\DeclareMathOperator{\clip}{clip}
+\DeclareMathOperator*{\argmin}{arg\,min}
+\newcommand{\bw}{\mathbf{w}}
+\newcommand{\balpha}{\boldsymbol{\alpha}}
+\newcommand{\R}{\mathbb{R}}