"ValueError: Weights is None" for Efficient Frontier Plot

[PooratPython]

I'm trying to generate an Efficient frontier plot but continue to get the same error - "ValueError: Weights is None".

The code I have used is as follows:

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import copy
from pypfopt import risk_models
from pypfopt import expected_returns
from pypfopt.efficient_frontier import EfficientFrontier
from pypfopt import plotting

stock_prices = pd.read_excel(r'C:\Users\fergaloconnor\Dropbox\UCC\FI6003\python\small_portfoliostocks.xlsx', 
                     parse_dates = ["date"], 
                     index_col = "date")
stock_prices = stock_prices.dropna()
stock_prices.describe()

mu = expected_returns.mean_historical_return(stock_prices)
Sigma = risk_models.sample_cov(stock_prices)
ef = EfficientFrontier(mu, Sigma)

fig, ax = plt.subplots()
ef_max_sharpe = copy.deepcopy(ef)
plotting.plot_efficient_frontier(ef, ax=ax, show_assets=True)

# Find the tangency portfolio
ef_max_sharpe.max_sharpe()
ret_tangent, std_tangent, _ = ef_max_sharpe.portfolio_performance()
ax.scatter(std_tangent, ret_tangent, marker="*", s=100, c="r", label="Max Sharpe")

# Generate random portfolios
n_samples = 100
w = np.random.dirichlet(np.ones(ef.n_assets), n_samples)
rets = w.dot(ef.expected_returns)
stds = np.sqrt(np.diag(w @ ef.cov_matrix @ w.T))
sharpes = rets / stds
ax.scatter(stds, rets, marker=".", c=sharpes, cmap="viridis_r")

# Output
ax.set_title("Efficient Frontier with random portfolios")
ax.legend()
plt.tight_layout()
plt.savefig("ef_scatter.png", dpi=200)
plt.show()

I used a few years of price for General Electric, JP Morgan Microsft and P&G daily - attached. I get the following error message and a blank graph:

ValueError                                Traceback (most recent call last)
<ipython-input-9-db7baad5d74e> in <module>
      1 fig, ax = plt.subplots()
      2 ef_max_sharpe = copy.deepcopy(ef)
----> 3 plotting.plot_efficient_frontier(ef, ax=ax, show_assets=True)
      4 
      5 # Find the tangency portfolio

C:\ProgramData\Anaconda3\lib\site-packages\pypfopt\plotting.py in plot_efficient_frontier(opt, ef_param, ef_param_range, points, ax, show_assets, show_tickers, **kwargs)
    261             ef_param_range = _ef_default_returns_range(opt, points)
    262 
--> 263         ax = _plot_ef(
    264             opt,
    265             ef_param,

C:\ProgramData\Anaconda3\lib\site-packages\pypfopt\plotting.py in _plot_ef(ef, ef_param, ef_param_range, ax, show_assets, show_tickers)
    194             )
    195 
--> 196         ret, sigma, _ = ef.portfolio_performance()
    197         mus.append(ret)
    198         sigmas.append(sigma)

C:\ProgramData\Anaconda3\lib\site-packages\pypfopt\efficient_frontier\efficient_frontier.py in portfolio_performance(self, verbose, risk_free_rate)
    443             risk_free_rate = self._risk_free_rate
    444 
--> 445         return base_optimizer.portfolio_performance(
    446             self.weights,
    447             self.expected_returns,

C:\ProgramData\Anaconda3\lib\site-packages\pypfopt\base_optimizer.py in portfolio_performance(weights, expected_returns, cov_matrix, verbose, risk_free_rate)
    537         new_weights = np.asarray(weights)
    538     else:
--> 539         raise ValueError("Weights is None")
    540 
    541     sigma = np.sqrt(objective_functions.portfolio_variance(new_weights, cov_matrix))

ValueError: Weights is None

Thanks in advance for your help!

[Suhetu]

I'm getting the same error, any luck figuring it out?

[giordanocolombi]

+1

[Suhetu]

So, here's what I observed in my case. When the annual returns are turning out to be negative or when the solver is unable to optimize the solution, that is what's causing this error.

[giordanocolombi]

@Suhetu look at #488