Computing returns
Every week, we will reinvest the entirety of our portfolio. Because of this, the return on the investment over the week will be reflected simply by the percentage change in the DJIA between the close of the first Monday and the close of the following Monday, but with the factor taken into account on whether we went short or long.
We have already accounted for going short or long using -1 or 1 for the signal, respectively. Now, we just need to calculate the percentage change, shift it by one week back in time, and multiply it by the signal value. We shift the percentage change back one week as we want to multiply the signal value for the current week by the next percentage change from the next week:
In [21]: base['PctChg'] = base.Close.pct_change().shift(-1) ...
Get Mastering pandas for Finance now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
