minimize2.py

import numpy as np
from scipy.optimize import minimize

w0 = 0
w1 = 0
w2 = 0
w3 = 0

ws = np.array([w0, w1, w2, w3])

## Constraints ################################################################

# sum of the weight should be 1
cons = [
    {'type': 'eq',   'fun': lambda x:  np.array([1 - np.sum(x)])}
]

# # All weights should be >= 0 and <= 1
# for i, w in enumerate(ws):
#     cons.extend([
#         {'type': 'ineq', 'fun': lambda x: np.array([ws[i]])},
#         {'type': 'ineq', 'fun': lambda x: np.array([1 - ws[i]])}
#     ])


## function to minimize #######################################################
tir = np.array([0.04, 0.045, 0.05, 0.0375])
initial_guess = np.array([0.1, 0.2, 0.5, 0.2])

bounds = [(0, 1), ] * len(ws)


def tir_portfolio(weights):
    return (-1) * np.sum(weights * tir)

## Let's minimize!!  ##########################################################

res = minimize(
    fun=tir_portfolio,
    x0=initial_guess,
    bounds=bounds,
    method='SLSQP',
    constraints=cons,
    options={'disp': True}
)

print(res)