25C here at 3pm, starting to warm up. Forecast is 36C for Saturday, looks like it will be a hot Christmas.
Unseasonably warm in West of Scotland.
Today’s low 10C (50F)
Today’s high 14C (58F)
Cloudy with some drizzle
Record high December temp (17C 65F) forecast for parts of Scotland.
Snowing here in Colorado! And high of 17 today, brrrr.
We got a light dusting of snow earlier this week, and that’s all it was. Long gone now.
33 degrees grey and windy in the Ozarks of Arkansas. First winter blast on the way 17 degrees and a dusting of snow. Unlike @rgcainmd I hate snow
View from my office
During the winter I sometimes see Bald Eagles on the far tree line. That white head sticks out like a sore thumb!
We see them around here, too, occasionally.
Current view from my office. Monday’s snow is long gone.
We’re finally getting some snow in N. Idaho. About four inches on the ground with 12-18 more inches coming in the next few days at my elevation It’s been a long time coming (we’ve had four winters in a row with far less than average snow, and poor snowpack in the mountains as well).
*edit: actually just got updates. We’re sub-zero © here for the next 10 days or so, and there is a minimum of 12" and a maximum of 28" forecast for my neck of the woods over the next 6 days. I’m sure it’ll change about twenty more times (weather forecasting is almost impossible more than 90 hours in advance with current computational models and equipment) between now and then, but I sure could use more than two feet of snow!
“The economy depends on economists to the same degree that the weather depends on meteorologists.”
Interestingly, economists and meteorologists run into the same problem when trying to make sensible forecasts based on mechanistic process-based rather than statistical-descriptive models. The statistical models are always pretty good, all things considered:
“December is likely to be cooler on average than June more than 30 degrees N of the Equator.”
“National economic growth tends to be correlated with positive growth in national stock markets.”
The problem is in trying to get detailed about complex systems. The best weather system mechanistic models, in particular, demonstrate chaotic mathematical behavior after t+96h. Which means it is impossible to predict with those models what will happen, but the same models can be used to explain past conditions from present system state
Same goes for economics… “hindsight is 20/20” really should be the professional slogan of The National Econometrics Guild (or whatever Economists’ professional organization is called)…
Even then, two different economists can and do look at the same data and draw different conclusions. It’s the inherent problem with any statistical data. As someone said, “There are three kinds of lies: lies, damned lies, and statistics.” I’ve seen that credited to at least a half dozen different people; my personal favorite is Disraeli. But regardless of who said it, it states a basic reality. Even the most honest, well intentioned experts can look at a statistical universe, view it from different mental contexts, and arrive at different conclusions as a result. Caveat emptor, yet again.
Hrmm, I’m not sure I’d blame it on the “statistics” part of the profession, myself. Statistical modeling tends to be very accurate regardless of discipline. The breakdown in statistics with the “lies, lies, and more lies” comes from interpretation and misunderstanding (or purposefully misrepresenting) statistical error.
The kind of economists who are most known for royally screwing things up in their forecasts (i.e., the ones so commonly asked to weigh in on politics in the US and Europe) are mechanistic, rather than statistical, modelers. They forecast, authoritatively, future economic conditions based on current economic state and their particular models of what different policy and market forces will do. They are, invariably, wrong. And deserve all the crap they get from more or less everyone I’m not even sure why we keep asking them what they think: we really don’t want to know.
The statistical modelers are the ones you don’t read about in the papers too much. They tend to describe broad trends based on statistical forecasting techniques that are pretty-well honed. They have big error bars and don’t make precise estimates of future parameters. Kind of like climate forecasting (which is actually pretty solid, as a discipline, from a mathematical and scientific standpoint).
I think you missed my point. If you scroll back and look, I was responding to your reference to after-the-fact interpretation, not future forecasting. And in any case, it’s not the accuracy or precision of the figures I called into question, but rather the (human) interpretation of them.
Two economists can look at the same graph or table and voice completely opposite conclusions about what they mean. I’ve seen it (literally) hundreds of times. I stand by what I said.
I wasn’t really criticizing what you were saying: I think we’re both making valid, if slightly different, points. I’m just suggesting that process modeling leading to forecasting of complex, non-linear systems (i.e., weather and national-scale economics) is impossible to do in a mathematically coherent fashion. You are right that two economists tend to interpret everything differently. But that is generalizable to more than economists. Get any two experts looking at the same material and they’re almost guaranteed to interpret things in opposite fashion. Probably a side-effect of telling people they’re experts!
And I agree with your general point about interpretation of statistical results. However, as someone that uses statistics on a daily basis to do my work, I think statistics and statisticians get a bad rap! The tools are great. It’s the pundits (and economists) you have to be wary of…
Oh, I can’t quarrel with any of that. But if my memory isn’t deceiving me (again), I think it was my economist joke that started all this.
Well, I’ll defer to you then! I thought we got off on this tangent because of my snide aside about the notorious unreliability of weather forecasters. Who may even be more disreputable than economists.
Actually, I think you’re right. I said my memory wasn’t any good. About as dependable as the two groups we’re discussing.