/jp/ - Otaku Culture

>>16458445
The update to the data might say otherwise.
A new post was made by the thread starter, and the OP got updated as well. Translation:

I talked with 梦路私信 and got permission to use their data.
After that I read the metadata and their test dates and fixed the hypothetical formulas, giving pic related

Grey background represents pre-fix data, it's invalid as you can see

Orange background represents data whose source can't be found, I suspect it's because シュン's page went down, due to the network being locked

For hypothesis 1's data, the test data with 38cm Kai is all marked in red, meaning that the theoretical value isn't within the 95% confidence interval, so at the level α = 0.05, the hypothesis is rejected.
But for hypothesis 1, when the number of 35.6cms is 0, it's actually the existing formula, when this part of data has a problem it implies that there's a problem with the original formula, and since it's a large part of the data clearly has disparities, it has to be concluded that during the 2016 June 1 maintenance, the fix wasn't just for 35.6cm series

Hence we have a new hypothesis 2, for where the June 1 maintenance didn't just affect 35.6cms, but the entire group 4, as to whether hypothesis 2 is correct, more tests are needed