In Step 2, we downloaded price data from Yahoo Finance, and only kept the adjusted close price for the returns calculation.

To calculate the simple returns, we used the pct_change method of pandas Series/DataFrame, which calculates the percentage change between the current and prior element (we can specify the number of lags, but for this specific case, the default value of 1 suffices).

To calculate the log returns, we followed the formula given in the introduction to this recipe. When dividing each element of the series by its lagged value, we used the shift method with a value of 1 to access the prior element. In the end, we took the natural logarithm of the divided values by using np.log.