When doing numerical operations on the headers, the strings get converted
to zero... To ignore this problem, remove the first line:

    cat netflix.tsv | sed 1d ...

You can also align the output columns with:

    ... | column -t

-------------------------------------------------

* only print lines between February 29, 2016 and March 4, 2016

    cat netflix.tsv | awk '$1 >= "2016-02-29" && $1 <= "2016-03-04"'

That might _possible_ with regular expressions, but probably not fun.


* sum the volumes for all days of January 2016

    cat netflix.tsv | awk '/^2016-01/ {sum+=$6} END {print sum}'

That's a recurring pattern: collect data and report.


* average the closing price over all days of March 2015

    cat netflix.tsv | awk '/^2015-03/ {sum+=$5; count++} END {print sum/count}'

Same pattern. Try your hand at standard deviation :-)


* check that all lines have 7 columns

    cat netflix.tsv | awk 'NF != 7'

The _header_ line has 8 columns and will print. It's good to check these things.

    cat netflix.tsv | awk '{print NF}' | sort -n | uniq -c

This will check how many columns there are and report each count.


* only print every other line (say, even lines)

    cat netflix.tsv | awk 'NR % 2 == 0'

NR is often used for things like this.


* remove empty lines in a file

    cat netflix.tsv | awk '/^$/ {next}; {print}'

That's probably the most explicit way of accomplishing that. Do lines with
blank spaces count as empty?

    cat netflix.tsv | awk 'NF'

Empty lines don't have fields ... making NF == 0 ... which is false.
Conversely, any non-blank line will be true and print.