By Lambert Strether of Corrente.
Bird Song of the Day
Snow Partridge, West Kameng, Arunachal Pradesh, India.
Readers, I hope your Christmas Eve and Day were everything you expected, and what you expected was good (and what you hoped for). The week between Christmas and New Year’s tends to be slow, and so my Water Cooler coverage this week will be a little erratic, perhaps confined to a single topic. (Tomorrow, however, I will do a complete post, just to prevent the pandemic(s) from getting away from me.)
Today’s topic is “proxies,” which are not the same as a voting proxies, except tangentially. My first step was naturally to check my OED, but it contained nothing to the purpose:
(Although the ecclesiastical usage is interesting; one wonders whether the Scholastics, say, had a theology of procuration.) The more fashion-forward American Heritage Dictionary does, highlighted:
(This is similar in structure to a market proxy.) I would refine the AHD’s definition as follows:
An entity or variable used to model or generate data assumed to resemble the data associated with another entity or variable that is typically more difficult to research, .
One obvious example of “unavailable or corrupt” data is Covid case counts for the United States. (Since wastewater data leads case data, it’s an OK proxy, even if the CDC’s national coverage is spotty. CDC does, at least, have county level data for most major airports, so at least I can highlight potential danger spots for readers who travel.) Here is a splendid example from The Economist of a proxy for actual, gen-u-wine, scientific case data, the sort of data public health establishment worthy of the name would be expected to collect, if only to help its “customers” with their personal risk assessments, but never mind that:
As readers know, loss of smell is one symptom of Covid. So it makes sense that Amazon reviews of “Yankee Candles” complaining that “this candle has no smell” would be a good proxy for Covid cases. And so it proved, in the study cited by the Economist. Now, I am but a humble tapewatcher, but if you think we have a good proxy here, it follows that we are in the midst of a surge, and it’s a bad one, on a par with Biden’s Omicron jouissance; I have drawn a grey “Fauci” line to show this[1]. Further, before I stopped running case data, I qualified it by saying that the real figures were much, much higher; I used a factor of six. If “no smell” is a good proxy, it looks like I was in the ballpark. Finally, we have seen a “divergence between clinical case curves and wastewater curves” starting in Q1 2022; here we see a similar divergence but for case curves and “smell” curves. Since this divergence is likely to be policy-driven, I have drawn a Democrat-blue line to show this.) Now, I’m not completely sure I accept the “no smell” proxy for Covid — is it confounded by flu, or RSV? — but it certainly underlines the message to “stay safe out there.”
NOTE
[1] If so, I was too conservative. On December 16, I rather than call a surge, I wrote:
I am calling a “Something Awful.” It’s gonna be bad, in some new way, and we don’t know how, yet.
Of course, we could some “Something Awful and a surge at the same time! Christmas data will tell us much.
Here is a second proxy, devised by GM, which I think wins a brilliancy prize. China has set about making its case data as corrupt as our own. What to do? Past proxies of cases have included backed up crematoria and crowded hospital parking lots (these can be determined from aerial photography). But this is great:
I’ve found that in the complete absence of data, a reliable proxy for what is happening is Wikipedia’s list of people with Wiki pages who died recently:
https://en.wikipedia.org/wiki/Deaths_in_2022
E.g. when Iran was having its first covered up outbreak in early 2020 it was very notable how many generals and politicians suddenly were dying at the same time.
There actually aren’t all that many Wiki pages for Chinese people despite the huge population (Western bias and so on), so usually it is 0 or 1 a day on that list. But here is what it looks like for the month of December:
Date/Number deaths on the page
25 4
24 1
23 5
22 3
21 2
20 2
19 1
18 1
17 1
16 1
15 2
14 0
13 0
12 0
11 0
10 0
9 0
8 0
7 0
6 1
5 0
4 0
3 0
2 0
1 1(they are still partial now).
Indeed!
Readers, can you suggest some good proxies? Not necessarily for Covid, but using something from the material, “real” world as a proxy or check on official data for, say, inflation or unemployment. Or manufacturing. Or heating fuel prices. Or, I suppose, metrics of any kind. Enjoy!
Contact information for plants: Readers, feel free to contact me at lambert [UNDERSCORE] strether [DOT] corrente [AT] yahoo [DOT] com, to (a) find out how to send me a check if you are allergic to PayPal and (b) to find out how to send me images of plants. Vegetables are fine! Fungi and coral are deemed to be honorary plants! If you want your handle to appear as a credit, please place it at the start of your mail in parentheses: (thus). Otherwise, I will anonymize by using your initials. See the previous Water Cooler (with plant) here. From JB:
JB writes: “My experiment with converting my pool to a 20k gallon aquarium, continues. The vegetation is now blooming, and not just the lily pads as I’ve shared in the past. What’s more, the Koi are growing fat and sassy. Everybody told me this wouldn’t work. Me? I’m thinking empirical results rule, or as Yogi Berra unforgettably said: ‘In theory there is no difference between theory and practice – in practice there is.’” Words to ponder!
Readers: Water Cooler is a standalone entity not covered by the annual NC fundraiser. So if you see a link you especially like, or an item you wouldn’t see anywhere else, please do not hesitate to express your appreciation in tangible form. Remember, a tip jar is for tipping! Regular positive feedback both makes me feel good and lets me know I’m on the right track with coverage. When I get no donations for five or ten days I get worried. More tangibly, a constant trickle of donations helps me with expenses, and I factor in that trickle when setting fundraising goals:
Here is the screen that will appear, which I have helpfully annotated:
If you hate PayPal, you can email me at lambert [UNDERSCORE] strether [DOT] corrente [AT] yahoo [DOT] com, and I will give you directions on how to send a check. Thank you!
