There

$3$

$$DataFrames:

$($ customer id age

$0114971\mathrm{.}0$

$175021\mathrm{.}0$

$242836\mathrm{.}0$

$336038\mathrm{.}0$

$481651\mathrm{.}0$

$........$

$862119353\mathrm{.}0$

$863115562\mathrm{.}0$

$8642123\mathrm{.}0$

$86546828\mathrm{.}0$

$86656222\mathrm{.}0$

$[867$ rows x $2$ columns$],$

customer$\_$id date shop$\_$id total$\_$price

$00\mathrm{.}02018-05-0440289\mathrm{.}673291$

$10\mathrm{.}02019-01-26401316\mathrm{.}898901$

$20\mathrm{.}02018-11-1842159\mathrm{.}414367$

$30\mathrm{.}02018-05-1242808\mathrm{.}791649$

$40\mathrm{.}02019-04-20421207\mathrm{.}855038$

$...............$

$40397$ NaN $2018-05-0537169\mathrm{.}086474$

$40398$ NaN $2018-09-297310\mathrm{.}993395$

$40399$ NaN $2018-10-31117\mathrm{.}535630$

$40400$ NaN $2019-02-1637218\mathrm{.}576020$

$40401$ NaN $2018-12-1549557\mathrm{.}510380$

$[40402$ rows x $4$ columns$],$

shop$\_$id offer$\_$begin offer$\_$end

$002019-02-282019-03-13$

$102018-07-222018-08-10$

$202018-09-212018-09-27$

$31$ NaN NaN

$42$ NaN NaN

$...........$

$178722018-05-072018-05-26$

$179732019-05-082019-06-11$

$180742018-09-142018-10-12$

$181742018-12-082019-01-10$

$182742018-08-222018-08-28$

$[183$ rows x $3$ columns$])$

Do people tend to spend more money when they make a purchase during a special offer? Let's check it$!$

First of all, we know that during December people spend more money in shops due to holidays and at the time shops usually announce more special offers during December. As we are interested in the effect of special offers alone, let us remove all purchases that happened during December.

Then create a table which rows are customers that contains two columns: in one column we have average of all prices of purchases that were made during special offers, and in the second the same for purchases that were not made during special offers. Drop all rows that contain NaN's. Then apply an appropriate statistical test to check if there is any statistically significant difference between these columns. Use two$-$sided alternative.