The Return, Part 3
Today I went to the Entry-Exit Bureau to submit my residence permit renewal. It is such a beauty to see a building with almost no people. The only point where there are people are those entering through the main entrance. They have cordoned the big area around the entrance to let people fall in line. As always, not a lot of people respect falling in line and the 一米 apart guideline and it is unfortunate that the entrance was guarded by only one lady.
She had to take temperatures so that it could be registered at another desk and she had to give guidance to everyone (locals and foreigners) about the procedures. I think she was overwhelmed because many do not respect the sanctity of the queue. Does not mean if there is a gap in the line, that you should fill the gap. Or in more salty language: Not every hole should be dicked. I think I just shouted unconsciously: 你们是不会排队吗? Can't you fall in line? I got a mix of reactions: an old fart who pretended not to hear (typically old fart strategy), a mid-age woman who stepped back, and the lady guarding the entrance finally speaking up and telling them to fall in line. Wrath is one of my favorite deadly sins.
The wrath never ends. After getting back home, we found that the kitchen window, which was left open (勤通风 as they say), allowed some construction dust to enter our home. Remember that construction/re-model two floors up. No fruit basket again and all we have is dust from drilling on our plates, kitchen utensils, and cooking pots. The funny thing is that the owner actually owns not just the flat being remodeled and the flat opposite it, but also another flat four floors down from the two flats!!!! So much money, but so much awfulness. The wrath never really ends.
I also was trying to organize files again and stumbled on PDFs of the crossword section of China Daily. I have a subscription for China Daily just so I could answer the crossword. The crossword section is part of their Life Fun page. It has some comics, other puzzles and an English-Chinese section where they teach you words from the news and other curiosities. It is one of the most useful pages in that newspaper. I discovered China Daily on a flight and was hooked on the crosswords because they get it from the New York Times! It is quite amazing. Regrettably, when I renewed my subscription last December 2019, the crosswords stopped effective January 2020. I got screwed. So, I am going through the entire back catalogue and downloading every crossword from 2013 onwards. That should sustain me for some time once I am off the “tenure track”. If you are interested in subscribing to China Daily, it is relatively easy but their website is not the most secure (so use passwords orthogonal to your most precious passwords). Once subscribed, you can download PDFs of the paper but only by page (as far as I can figure out). So I focus on only one page: Life Fun. Sometimes, I do encounter Life Reading, which does have some approved literature.
What else did I do today? Oh yes, I am about halfway through The Elements of Data Analytic Style (clearly a nod to Strunk and White) by Jeff Leek. It is quite a short book for anyone trying to do data analysis. It has a very nice flowchart at the beginning of the book differentiating the types of data analysis, which I have found useful for admin purposes, especially when thinking about the thesis writing process. Apparently, there are 6 types: descriptive, exploratory, inferential, predictive, causal, and mechanistic. In my opinion, most master theses I have encountered belong to inferential type of data analysis. It is rare to see predictive data analysis but even rarer to see causal data analysis. One thing that was curious about the flowchart is that the phrase “policy implications” never show up. This phrase also does not show up in the entire book! The word “motivation” shows up once as part of the written analysis. Another very interesting point (that we take for granted, and probably should let students know) is that generating figures for oneself and for others are two very different things. You can download the book for free but you can also pay as you want. If you are able to pay, do give some money. Not sure if it is tax-deductible but that should not be the reason to give money. Don't be an asshole like the Zuck.
I was also catching up on some other reading. A new master's student wants to study overdifferenced time series for the master's thesis. I think this was influenced by the student taking my Advanced Econometrics 2 class where we took some time to read Nelson and Plosser (1982) in its entirety. The interest is to determine whether or not there was overdifferencing in an analysis. The student was able to prove, as part of a small project, that the OLS estimator of a regression coefficient of interest using the data on the overdifferenced series is inconsistent for the true regression coefficient (in a simple setting). The student then develops a bias-corrected estimator but requires a lot of strong assumptions. The student also proposed a test for the “presence” of overdifferencing, but his Monte Carlo indicates that it does not work. A nice and early trial to find out how far one could take an analysis.
I never imagined that I would read as much time series papers as I do now. I never liked applied time series analysis before, perhaps it may be because of how cheap (from a software perspective) it has been to produce results. All this beautiful theory in time series somehow got tainted by the improper use of causality tests, Johansen cointegration tests, and many more. In my classes, I actually teach Granger causality testing while doing F-tests and I use the example by Thurman and Fisher (1988?) on which came first, the chicken or the egg? Just to poke fun at this whole G-causality business.
I still have a couple of things to figure out for this blog. I plan to write more econometrics related stuff. I also want to enable comments. I would want to enable math input here. I thought I already figured it out but I have to give this some more time. In principle, I should be able to use LaTeX code here. I also want to start writing a book (after teaching econometrics-related courses to a variety of audiences for many years) called The Applied Theory of Econometrics, a nod to Deirdre McCloskey's The Applied Theory of Price, back when she was The Donald. If I manage to set comments up, perhaps the book can become a living document.