Disliked{quote} No, I used Java (and Apple Numbers for the plots). For these simple experiments probably any language will do, although admittedly Python is very convenient for this kind of scripting. However, I do most of my ML using proprietary algorithms that are not available in the Python libraries. Implementing them in Python would be too slow and implementing them in C as a loadable Python library is too much hassle for my taste. Hence I do mostly work with Java which nowadays is as fast as C++ (without the complexities of writing robust C++ code)....Ignored
Disliked{quote} Ah, I see. That is different from both methods I tried. Since we are looking at seasonal effects on a daily basis, it is very important the matching is exact. However, here it matters what seasonal event we are trying to find. Certain holidays are always at the same calendar date (and therefore not on the same day of the week). Other events are always on the same day of the week but on different calendar dates (e.g. returning to office on Monday after holiday, anticipating an upcoming holiday on Friday, etc.). So it is hard to tell which...Ignored