Monopolist’s Effective Supply Function

Randall Romero Aguilar, PhD

This demo is based on the original Matlab demo accompanying the Computational Economics and Finance 2001 textbook by Mario Miranda and Paul Fackler.

Original (Matlab) CompEcon file: demapp10.m

Running this file requires the Python version of CompEcon. This can be installed with pip by running

!pip install compecon --upgrade

Last updated: 2022-Oct-22


Initial tasks

import numpy as np
import matplotlib.pyplot as plt
from compecon import BasisChebyshev, NLP

Residual Function

def resid(c):
    Q.c = c
    q = Q(p)
    marginal_income = p + q / (-3.5 * p **(-4.5))
    marginal_cost = np.sqrt(q) + q ** 2
    return  marginal_income - marginal_cost 

Approximation structure

n, a, b = 21, 0.5, 2.5
Q = BasisChebyshev(n, a, b)
c0 = np.zeros(n)
c0[0] = 2
p = Q.nodes

Solve for effective supply function

monopoly = NLP(resid)
Q.c = monopoly.broyden(c0)

Plot effective supply

nplot = 1000
p = np.linspace(a, b, nplot)
rplot = resid(Q.c)
fig1, ax = plt.subplots()
ax.set(title="Monopolist's Effective Supply Curve",
       xlabel='Quantity', 
       ylabel='Price')
ax.plot(Q(p), p);
../../_images/10 Monopolist's Effective Supply Function_11_0.png

Plot residual

fig2, ax = plt.subplots()
ax.set(title='Functional Equation Residual',
       xlabel='Price',
       ylabel='Residual')
ax.hlines(0, a, b, 'k', '--')
ax.plot(p, rplot);
../../_images/10 Monopolist's Effective Supply Function_13_0.png