Software

Open Source Software

Abstract: The Neer Match package provides tools to generate and fit Artificial Neural Networks for entity/record matching using neurosymbolic learning based on customizable similarity encodings. The similarity encodings instruct the constructed models on associating multiple fields from two datasets using similarity functions. The package offers a machine learning-driven probabilistic matching process extending uni-dimensional fuzzy matching that can intrinsically reason about its matching recommendations. The matching approach can be easily adjusted and allows embedding domain knowledge for various entity-matching settings. Keywords: Entity matching, Entity resolution, Database linking, Machine learning, Record resolution, Similarity encoding
Abstract: The structural-schooling repository provides a collection of modular scripts used to calibrate an economic model of technological structural changes with households comprised of two individuals with distinct genders with endogenous schooling choices. The model is semi-analytically solvable, and the optimization procedure for approximating its solutions is nested in the calibration procedure. The model is part of the work on gender schooling differences by Karapanagiotis & Reimers (2024). It combines the structural change elements of the model of Ngai & Petrongolo (2017), the educational choice elements of Restuccia & Vandenbroucke (2014), and the seminal work of Kongsamut, Rebelo, & Xie (2001) on non-homothetic preferences. Keywords: Development, Education, general equilibrium, Structural change
Abstract: Provides estimation methods for markets in equilibrium and disequilibrium. Supports the estimation of an equilibrium and four disequilibrium models with both correlated and independent shocks. Also provides post-estimation analysis tools, such as aggregation, marginal effect, and shortage calculations. The estimation methods are based on full information maximum likelihood techniques given in Maddala and Nelson (1974) <doi:10.2307/1914215>. They are implemented using the analytic derivative expressions calculated in Karapanagiotis (2020) <doi:10.2139/ssrn.3525622>. Standard errors can be estimated by adjusting for heteroscedasticity or clustering. The equilibrium estimation constitutes a case of a system of linear, simultaneous equations. The disequilibrium models, instead, replace the market-clearing condition with a non-linear, short-side rule and allow for different specifications of price dynamics. Keywords: Disequilibrium, full-information-maximum-likelihood, market-clearing
Abstract: The code implements a simple version of the radial attention model and solves it. The solver is based on a concurrent re-formulation of the value function iteration algorithm. Further, it uses an adaptive search grid implementation to provide more accurate optimal control approximations. For more implementation details see the documentation. Keywords: attentional costs, endogenous choice sets, learning, radial attention, value function iteration

Proprietary Software

Abstract: Database-driven Business Management Software. Budget and accounting management. Sole proprietorship and consulting solution.