You have access to sales data of 30000 stores of a Fortune Global 500 company. You are interested in finding if there exist a significant change in overall sales in year 2019 in comparison to year 2018. Because you want to keep the analysis relatively simple, you pick 1000 stores at random and conduct a two-sample hypothesis test comparing overall sales in 2019 with overall sales in 2018 for these stores. To make sure the result is reliable, you should repeat this process an arbitrarily large number of times, and take the smallest p-value as your final p-value