Regarding the calculations of the Stressed scenario for Category 2 discussed in Annex IV in the delegated regulation 2017/653 (https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A02017R0653-…;
Mainly the confusion comes from paragraph 11, which says “…where zα is a proper selected value of the PRIIP at the extreme percentile that corresponds to 1 % for 1 year and to 5 % for the other holding periods.”. For all other scenarios the values for zα have been given (they are each scenarios respective N(0,1)-percentiles). In “PRIIPs – Flow diagram for the risk and reward calculations in the PRIIPs KID” (Flow diagrams for the risk and reward calculations.pdf (europa.eu)), Section 4 Part 4 b), they use N(0,1)-percentiles for zα in the stressed scenario as well. Though when reading paragraph 11 directly after having read paragraph 10 (d), where one takes percentiles from a calculated sample set rather that a N(0,1)-distribution, it made me think one was supposed to use a sample set’s (the historical lognormal returns) percentiles for zα in paragraph 11 rather than the percentiles from a N(0,1)-distribution. 
Can you confirm that one indeed is meant to use the percentiles from a N(0,1)-distribution for zα in paragraph 11 as is done in the “PRIIPs – Flow diagram…”-file?