CasseySmithCV

Produces critical values using the methods described in Cassey Smith (2014). Code is modernized from version published with paper to take advantage of advances in computing.


Keywords
EG, Statistic, Monte, Carlo, Statistics
License
MIT
Install
pip install CasseySmithCV==0.1.0

Documentation

CasseySmithCV

Simulation software to generate critical values for the Ellison-Glaeser index (1997). This is updated software from Cassey and Smith (2014); it modernizes the codebase and increases the performance of the simulations.

Examples

Run a simulation of industries with 20 plants and sigma value of 1.

from CasseySmithCV import runSimulation, SimulationTable, SaveSimulationTable
resultDictionary = runSimulation(20,1)

Run all combinations of [20, 30] plants and [0.5, 1] sigma values:

from CasseySmithCV import runSimulation, SimulationTable, SaveSimulationTable
resultDataFrame = SimulationTable([20, 30],[0.5, 1])

Run all combinations of [20, 30] plants and [0.5, 1] sigma values and save the results to a file:

from CasseySmithCV import runSimulation, SimulationTable, SaveSimulationTable
SaveSimulationTable('filename.csv', [20, 30],[0.5, 1])

Installation

This package can be installed by using either the pip or conda command:

Installing by pip

pip install CasseySmithCV

Installing by conda

conda install -c tazzben casseysmithcv