Eigen Portfolio construction via PCA (75/100) about machine-learning-and-reinforcement-learning-in-finance HOT 1 CLOSED

e97d5cd, managed to cleared Eigen Portfolios: Sorted by Sharpe Ratio by changing the calculation for annualized_return from

annualized_return = ts_returns.mean() * periods_per_year

to

n_years = ts_returns.shape[0] / periods_per_year
annualized_return = np.power(np.prod(1 + ts_returns),(1 / n_years)) - 1

However, results will have NaN value which I then used results.dropna(inplace=True) to remove the NaN value. This helps to cleared Eigen Portfolios: Sorted by Sharpe Ratio but Portfolio with the Highest Sharpe Ratio would fail the grader. This can be easily resolved by using

idx_highest_sharpe = np.nanargmax(sharpe_metric)

to obtained the idx with the highest sharpe instead of

idx_highest_sharpe = np.argmax(sharpe_metric).

from machine-learning-and-reinforcement-learning-in-finance.