Time to play another exciting round of “COVID Peak Calling!!!”
And who is our first contestant this morning? Why it’s…
We’re not buying it, but it’s not a bad estimate, by any stretch. The problem is what happens in India, Africa, and South America. In our informal “back of the envelope” work, by May 1, the global case count should be around 38.535-million. And U.S. deaths following could be closer to 135-thousand.
What’s more, since our mortality rate calcs are running 4.486% this morning, the global body count by then could be up around 1.729-million and the U.S. share (proportionately) ought to be around 75.065-million by May 1, but it could run another week or two from there and that’s where the higher estimate we cobbled up came from. Except, the data will be very noisy because “social distancing” is not an even thing.
Why, if everyone would stay home for the next five-years….
Out here in the East Texas Outback, where people tend to live on acreage and not bumping elbows in the elevators, compliance is easy-peasy. In Mid-Town, NYC, a little tougher and less convenient. That’s one reason the virus strike team’s looking at publishing a set of “County Risk” assessments to guide local responses.
As far as I know, locally, the Palestine Herald reports the closest cases to us are some 45-miles north up in Smith County (Tyler, TX) with seven cases; but none here yet. That is, officially.
The problem is that since we’ve been running numbers for my son up north (he’s a firefighter/EMT/safety officer) and plugging in our pop-data suggests we could have as many as 4.2 cases here right now. Although, that would infer we won’t get a death in Anderson County, Texas, until out in the Easter timeframe (April 12, for atheists.)
Ugly Multivariate Compounding
What makes assessing personal plans, contingencies, strategies, and tactics so difficult is that depending on (location, location, location) the number of variables involved in future-modeling just becomes mind-boggling.
It’s not unlike “flying right-seat” with the Wright Brothers first time up. You’re never sure is flying is doing to work out as your calculations indicate.
For most of this, there are three major sets of variables, if we put them into piles.
One pile is the supply line and resupply issues. That’s where the snakes are now.
For American-Made goods, shouldn’t be bad and that’s grand because we do most of our own food, save that which comes in from the areas south of the Rio Grande Valley (TGV). Where the supply chain is NOT robust is coming in from Asia. Due in large part to not knowing exactly how bad things got in China.
The Chinese case-count is continuing lower than the USA count, but I’d sure as hell trust CDC numbers before trusting the “Party-approved” ChiCom data. Oh, and the whole “Spy-vs.-Spy” angle of the world’s economic superpowers has to figure into the supply chain, too. As China is now closing itself off to foreigners (Mind Dynasty Replay, anyone?). And in order not to tip our hand on readiness, the U.S. dot-mil types aren’t talking about our troop’s health, either.
As a result, we are in a painful period of massive socio-economic readjustments. That “roaring twenties” lifestyle we’ve been yammering about? Going, going….
Foremost among which is you may notice the Internet being a little slower than usual. We’re doing our part by limiting our graphics and compressing more – which may drop the quality of images on your phone, but all an analog of “social distancing” on the data front.
Recalibrating Your Thinking
As a useful tool for deeper-thinking, we don’t just need to “collect facts” – we also need some reasonable means to assemble them into useable information. Remember those Business 603 classes? Boy, I sure do. Data is useless until it’s processed and assembled into information or knowledge which can then used to inform executive decision-making and action.
Much as I hate statistics, multivariate-thinking is like building a “coat rack for data” to hang on:
“Multivariate statistics is a subdivision of statistics encompassing the simultaneous observation and analysis of more than one outcome variable. The application of multivariate statistics is multivariate analysis.
Multivariate statistics concerns understanding the different aims and background of each of the different forms of multivariate analysis, and how they relate to each other. The practical application of multivariate statistics to a particular problem may involve several types of univariate and multivariate analyses in order to understand the relationships between variables and their relevance to the problem being studied.
In addition, multivariate statistics is concerned with multivariate probability distributions, in terms of both
- How these can be used to represent the distributions of observed data;
- How they can be used as part of statistical inference, particularly where several different quantities are of interest to the same analysis.
Certain types of problems involving multivariate data, for example simple linear regression and multiple regression, are not usually considered to be special cases of multivariate statistics because the analysis is dealt with by considering the (univariate) conditional distribution of a single outcome variable given the other variables.”
“How many trucks can move goods east out of Los Angeles on I-40?” is a kind of univariate problem. The freeway is x lanes wide and at 55 or 60, with normal spacing, there’s your limit.
On the other hand? How many trucks – filled to 80% of capacity, with adequate fuel and food enroute – along with operating warehouses on arrival – becomes a multivariate problem.
Or not. Because once you go down the multivariate rabbit hole, you pop into the simultaneous equations world:
“Simultaneous equation models are a type of statistical model in which the dependent variables are functions of other dependent variables, rather than just independent variables. This means some of the explanatory variables are jointly determined with the dependent variable, which in economics usually is the consequence of some underlying equilibrium mechanism. For instance, in the simple model of supply and demand, price and quantity are jointly determined.
Simultaneity poses challenges for the estimation of the statistical parameters of interest, because the Gauss–Markov assumption of strict erogeneity is violated. And while it would be natural to estimate all simultaneous equations at once, this often leads to a computationally costly non-linear optimization problem even for the simplest system of linear equations. This situation prompted the development, spearheaded by the Cowles Commission in the 1940s and 1950s, of various techniques that estimated each equation in the model seriatim, most notably limited information maximum likelihood and two-stage least squares.”
This is where people tend to get thrown under the statistical bus. The terms get complicated, so you need to put on your thinking cap and use mental modeling to seek the truth you’re looking for.
Let’s go back to the Trucking problem on I-40. We understand it to be a multivariate problem, but as such, any models we evolve will be subject to both feedforward and feedback.
Let’s take feedback first. Obviously, if our truck, loaded with goods fails to arrive, there are lower chances of workers being around the next time a truck shows up. Some may have starved to death, moved on, or whatever. That would be feedback. Cause followed by effect. Feedforward is effect followed by cause in some related area. Ummm…confused?
Feedforward is a little different in that the truck’s arrival triggers further future events in the arriving community. People get food, go back to other jobs, making toilet paper we hope, and so forth.
We’re in a Crisis of Complexity
Ultimately, this is why the stock market is undergoing wild gyrations this week.
Because, as in our simple trucking example, we have a virus and it will spread. Which is one set of variables. But, we know spread can be contained through social distancing because no hosts means no passing it on. Well, except for those asshole kids who went on spring break and then brought the disease back to the cities they came from…but that gives rise to yet-another variable.
Computationally, therefore, we have several ways to approach the problem. One method is to model the overall supply chain and some n number of subsets, all of which are interoperative (feeding forwards and back) in that you can’t run out of diesel for the trucks, but at the same time, you can’t run out of burgers for the drivers kind of thing. And every freaks about toilet paper…
We can flip back through history to see how past economic collapse event-chains fell apart. With (another) caveat that the rate of social breakdown and the ultimate workout will be different this time around.
Whew! Didn’t mean to “go deep” on you, but this is nothing compared to tomorrow’s Peoplenomics report (“Evidence of Cultural Knowledge in Entanglement”). In that one, we get down to the social substrate and “How Everything Works” level…up here, it’s back of envelope and pencils, please.
IRT (In Real-time) this morning’s short answer is “The puzzle answers MAY be seen by simply looking at a stock market chart comparing the present-day Aggregate of Market prices versus the Dow Jones Industrials which went through a similar propagating collapse period in 1929-1940.” Like so:
Since we’re conserving graphics overhead, the low we just went through (March 23rd) lines up with the 1929 era’s first low (325.17) on October 4 of 1929. And, of course, from there the ’29 market rallied back up to the 356.92 level which was set October 10, 1929.
There’s an argument that the market of present-day should not decline yet. We really ought to rally a bit more next week. However, if the “rally” is already over (we’re at the green circle in the chart based on early futures) then the low in the market could be set 35-calendar days from yesterday: Comes out April 30th.
On the other hand, if the market were to peak at the end of next week, April 3, then this would suggest a bottom low of the market (if we’re stuck in walking in the same ruts as the last collapse) around May 8th.
The bad news is all this? Well, there’s another tool I built a year or two back which is on the Peoplenomics subscriber site on the Master Index page over here. “Download: Brainamp.xls A spreadsheet to estimate 5 Elliott Waves from Wave 1 data points.”
We’ll go through that on the subscriber side tomorrow…it ain’t pr3etty and we’ll need barf bags for all hands.
But depending on whether the market can “put on more beans” next week, that’s a stab at bracketing where we MIGHT/MAYBE/COULD go in the near future. And that would be a retest of the 2009 market lows in early May. S&P in the 600’s.
Does It Really Matter?
I’d be derelict if I didn’t mention the Personal Income and Expenditures report just out from the Bureau of Economic Analysis: We read this like the stats from the wild excesses of 1929…Kinda like “How did our Roaring Twenties end?”
Personal income increased $106.8 billion (0.6 percent) in February according to estimates released today by the Bureau of Economic Analysis (table 3 and table 5). Disposable personal income (DPI) increased $88.7 billion (0.5 percent) and personal consumption expenditures (PCE) increased $27.7 billion (0.2 percent).
Real DPI Increased 0.4 percent in February and Real PCE increased 0.1 percent (table 5 and table 7). The PCE price index increased 0.1 percent. Excluding food and energy, the PCE price index increased 0.2 percent (table 9).
Not that it matters: Market futures were down more than 500 points on the Dow when I checked earlier. But, I’m expecting a big running of the shorts rally into the close if short-term traders all head for the exits for the weekend.
That’s almost as much fun as going to a slaughterhouse to watch the squeeze chutes…
Meantime, the article on ZeroHedge about why the “stimulus bill needs to be stopped” echoes many of our sentiments, we’re mindful that the fix is in and has been sitting there since last July when the democrats drafted HR 748…
And if you think the news lately is enough to drive you to drink? Go read Fortune’s The liquor industry faces an uncertain future. But if it survived Prohibition, it can survive the coronavirus.
We’ll try to “speak-easy” come Monday, A month from hell just ahead.
Write when you get rich,