November 2020 Newsletter

This is a monthly newsletter of PSL model releases, events, and updates.
Subscribe at https://www.pslmodels.org and follow us on Twitter @PSLModels.

New Initiatives

• PSL launched a "Demo Day" series where developers showcase their models. It has so far covered Tax-Brain (with Matt Jensen) and Tax-Cruncher (with Peter Metz). Subscribe to our YouTube channel and join the next event on November 30 where Jason DeBacker will demo Cost-of-Capital-Calculator.
• PSL also has a new blog for sharing major updates on our catalog. Drop our editor, Max Ghenis, a line to share your work or suggest content.
• For an overview of PSL, check out this Virtual Global Village podcast featuring Jason DeBacker.

microdf

A package for analyzing economic microdata as pandas DataFrames.

General updates

• Used in UBI Center's studies to show how universal basic income would affect veterans, to compare student debt cancellation to universal payments, and to simulate the effects of child allowances on poverty and inequality across US states.
• Max Ghenis gave a talk at PyData Global 2020 on building an open source think tank using PSL tools such as microdf.

ParamTools

A library for parameter processing and validation with a focus on computational modeling projects

Releases and new features

• Release 0.16.0  —  Improved API for querying parameter values and supporting custom data types.

General updates

• Presented at the PyData Global 2020.

PCI-China

A machine learning algorithm that predicts China's policy changes by mining its state-controlled media

Releases and new features

• Release 2020-Q3  —  Updates the PCI-China indicator to 2020 Q3, shedding light on China’s post-COVID policy priorities.

General updates

• Demonstrated at the Data Natives 2020.

Tax-Cruncher

A model that calculates federal tax liabilities from individual data under a range of policy scenarios

Releases and new features

• Release 0.6.0  —  Redefined input parameters (#53) and made minor modifications to the web application outputs (#51, #54).

General updates

• Used to create an app that calculates a taxpayer’s liabilities under the Biden tax plan. Check out the app housed by Compute Studio and an American Enterprise Institute blog post.

OG-USA-Calibration ^*

A library for estimating and calibrating the OG-USA model

General updates

• This is newly included as an incubating project at PSL!
• The library contains scripts to read in data sources and estimate/calibrate parameters used in the OG-USA model. Check out its documentation here.

PCI-Crackdown ^*

A machine learning algorithm that predicts crackdowns on protests by the Chinese military

General updates

• Demonstrated at the International Symposium on Forecasting.

• Behavioral-Responses is a partial-equilibrium behavioral module that works with Tax-Calculator.
• Cost-of-Capital-Calculator evaluates the effect of US federal taxes on businesses' investment incentives.
• microdf is a package for analyzing economic microdata as pandas DataFrames.
• OG-USA is an overlapping-generations model for evaluating fiscal policy in the United States.
• ParamTools is a library for parameter processing and validation with a focus on computational modeling projects.
• PCI-China is a machine learning algorithm that predicts China's policy changes by mining its state-controlled media.
• Tax-Brain is an integrator package for multiple open source tax models.
• Tax-Calculator is a USA federal individual income and payroll tax microsimulation model.
• Tax-Cruncher calculates federal tax liabilities from individual data under a range of policy scenarios.