First Article
November 22, 2024
Blog
Average Excess Returns: Calculate the time-series mean excess return for each portfolio and perform a 𝑡𝑡-test of the null hypothesis that the average excess return is equal to zero. The 𝑡𝑡- statistic from your 𝑡𝑡-test should be adjusted following Newey and West (1987) using 12 lags.
def newey_west_ttest(excess_returns, lags=12): excess_returns = excess_returns[~np.isnan(excess_returns)] if len(excess_returns) > 0: model = sm.OLS(excess_returns, np.ones(len(excess_returns))) results = model.fit(cov_type='HAC', cov_kwds={'maxlags': lags}) return results.params[0], results.tvalues[0] else: return None, None