Errata

Regarding to the original paper "Man vs. Machine: Comparing Discretionary and Systematic Hedge Fund Performance", https://www.readcube.com/articles/10.2139%2Fssrn.2880641, at the column for Systematic macro in the table on page2, the adjusted return volatility is 10.93%.

So it has to be corrected from 0.93% to 10.93%.

Item 6 says "Prints three data rows from every DataFrame object", however the output provided on pages 60-61 show 5 lines printed from every data frame.

The explanation for item 1 says "This generates a sample financial data set with 2,500,000 rows and five columns with a one second frequency; the sample data is rounded to two digits."
It should probably say "... the sample data is rounded to four decimal places".

The text says "The gross performance of the strategy with the original parameterization is 1.24 or 124%. The optimized strategy yields an absolute return of 1.44 or 144%...", whereas the output and plots suggest the performance of the original strategy is 1.29 and of the optimised strategy - 1.50.

The 1st sentence:
With regard to the special method __main__,

__main__ should be __init__?

The argument cols does not appear to be used when selecting the columns to plot in the plot_data function (it always plots the price).

Item 13 says "...a buy order for 0 is initiated"
Should it be instead "...a buy order for self.units * 2 is initiated"?

Similarly for item 19.

Item 6 says "This starts the streaming and therewith the trading; it stops after 100 ticks", whereas the code on page 241 seems to suggest the streaming stops after 500 ticks?

The logic of the example suggests it's meant to demonstrate using a variable-length argument, however the list is passed as a single argument, so the args variable in f will only contain a single value (which is the list object) and will print only one line.

For the list to be interpreted as separate arguments, it has to be passed as:
f(*l)
This will print five lines with one letter in each line.