/Recon

Tools to compute and visualize economic models

Primary LanguageRGNU General Public License v3.0GPL-3.0

Build Status GitHub GitHub release Downloads Rdoc

Recon - Computational Tools for Economics

This package aims to provide undergraduate students and professors realiable ways of computing and exploring canonical models in economics. It creates a rich enviroment for comparative statics, visualization of results and better understanding of model outcomes.

Installation

You can install the latest version of Recon by running:

devtools::install_github('pedrocava/Recon')

## or directly from CRAN

install.packages('Recon')

Which should return something similar to:

Installing package intoC:/Users/Pedro/Documents/R/win-library/3.5’
(aslibis unspecified)
* installing *source* package 'Recon' ...
** R
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
  converting help for package 'Recon'
    finding HTML links ... done
    MRW_steady_state                        html  
    cobb_douglas                            html  
    cobb_douglas_2                          html  
    cournot_solver                          html  
    grid2                                   html  
    monopoly_solver                         html  
    sim_mixed_nasheq                        html  
    sim_pure_nasheq                         html  
    solow_steady_state                      html  
    stackelberg_solver                      html  
** building package indices
** testing if installed package can be loaded
*** arch - i386
*** arch - x64
* DONE (Recon)
In R CMD INSTALL

Please note that you should have Rtools installed.

Features by theme

To get a general view of what's implemented in Recon directly from R you can run:

help(package = Recon)

Economic Growth

  • Steady state solution to the model presented in Solow (1956)
  • Steady state solution to the model presented in Mankiw, Romer and Weil (1992)

Cobb-Douglas Functions

  • Computing Cobb-Douglas functions with 2 inputs
  • Computing Cobb-Douglas functions with $n$-inputs

Miscellaneous

  • Grid generating function, so one can map results easily

Game Theory

  • Finds the Pure Strategies Nash Equilibrium of a 2 person simultaneous game
  • Finds the Mixed Strategies Nash Equilibrium of a 2 person simultaneous game, for 2x2 pay-off matrices

Imperfect Competition

  • Solution to a Cournot Duopoly Model, with linear and non-linear cost and demand curves
  • Solution to a Stackelberg Duopoly Model, with non-linear cost curves
  • Solution to Monopoly Profit Maximization

Upcoming Features and Suggestions

Feel free to ask/suggest a model for implementation. If you want to implement one youself, fork this repo and then do a pull request with your new functions. After reviewing, you'll be credited and your functions will be part of Recon. Features that are currently under work:

  • Bertrand Duopoly Model with differentiated goods
  • Cournot Oligpoly with n firms and homogenous cost functions
  • McCall Search Model
  • Optimal consumer choice
  • Optimal output for a single firm under perfect competition
  • Walrasian Auctions and primitive forms of Computable General Equilibrium

What we're aiming for in the long run

Right now Recon is just a small group of relatively simple functions. There's a lot to be made before version 1.0, but here's a taste of what's planned for it:

  • Classes representing economic entities such as consumers, firms and markets.
  • Methods for handling these classes and solving models with them
  • Rich variety of functions for visualization and comparative statics

Citation

To cite package ‘Recon’ in publications use:

Pedro Cavalcante Oliveira, Diego S. Cardoso and Marcelo Gelati (2019). Recon: Computational Tools for Economics. R package version 0.3.0.0. https://CRAN.R-project.org/package=Recon

A BibTeX entry for LaTeX users is:

  @Manual{recon,
    title = {Recon: Computational Tools for Economics},
    author = {Pedro {Cavalcante Oliveira} and Diego {S. Cardoso} and Marcelo Gelati},
    year = {2019},
    note = {R package version 0.3.0.0},
    url = {https://CRAN.R-project.org/package=Recon},
  }