Finding the World's Economic Center of Gravity

http://nbviewer.ipython.org/github/djv/world_economic_center/blob/master/map.ipynb

by jsm386 • 12 years ago

64 22 12 years ago

nbviewer.ipython.org

waps2• 12 years ago

The problem with this chart is that it's heavily biasing the "center of economic activity" towards the center of the Map. The reason it's constantly above Europe is that Europe is pictured between Asia and America.

If you center the Map over America you'll see the center of economic activity being America. If you center it over Asia, it'll be Asia.

Therefore this map does not contain much more information than "Greenwich lies in the middle".

dkokelley• 12 years ago

Agreed to a point. The "middle of the map" theory only applies if you assume economic activity is relatively evenly spaced (which it mostly is). Unfortunately this representation won't tell us much about where economic activity really is.

There are better ways to represent the information presented. A 3D map would put economic activity somewhere in the middle of the earth, but you would eliminate the "starting center" bias.

waps2• 12 years ago

3d would be pretty hard to see. But there's plenty of empty spots on the earth that are in the middle of all economic centers. Ideally bordering every economic center ... since economic activity (and land mass, which is likely why) is concentrated in the northern hemisphere, why not center the map around the north pole ? No economic activity, roughly equidistant of the 3 big centers. Sure it might not be perfect, as South Africa will effectively be counted with Europe and Australia with Japan, South America with the US (well, with California effectively), but it ought to be a bit better than this.

If that is still less than optimal, maybe try the south pole ?

Seems to me it might correct the bias seen here.

djvv3• 12 years ago

Author here. I agree with your point but the map is still useful for seeing the directional trend. I took this projection because it is the usual way of plotting the world on a 2d map.

philh• 12 years ago

the map is still useful for seeing the directional trend.

Isn't that also true of the original map, though?

It's not at all clear to me that your visualization is an improvement on the original. The center has moved from Scandinavia to Spain, and I can see why Scandinavia wasn't a great center, but I don't see why Spain is better. Meanwhile, your visualization depends on the projection used, which seems highly nonintuitive.

(If you go from the USA to China by the shortest distance, you go near the north pole. Calling the north pole the average of the two places may seem strange, but calling any other place the average is even stranger.)

analog312• 12 years ago

I'd try to resolve this by displaying the center of gravity in 3 dimensions. Then I could project each point to the surface if I wanted.

archgoon• 12 years ago

This was the Economist's original method.

From the article:

"There was one thing which bothered me, and I hope it bothers you too, the points from the 20th century are all positioned in or above Scandinavia which seems unlikely to be the center of anything in the world. You can find the reason for that in the caption on the McKinsey webpage. Their report looks at the Earth as a sphere and finds the economic center of gravity which falls somewhere inside the sphere. To plot it on the map, they take a radius through the center of gravity and intersect it with the surface."

djvv3• 12 years ago

That's the method from the original article. It distorts the ratio between the economies in the North vs. South hemispheres.

mbq• 12 years ago

Agree, COM in Mercator projection is even more wrong than the Economist method. If one wants to have it on sphere the only correct way to do this is by using the great circle distance. Obviously it still leaves a question what does this center of economic activity actually shows (;

codezero2• 12 years ago

This is interesting because I was thinking about a somewhat related sort of statistical measure (trying to determine average times over repeated measurements of a wall clock) and happened upon this: https://en.wikipedia.org/wiki/Mean_of_circular_quantities

I think this particular trend could benefit from some insights therein, they would help avoid the issue of hovering around the center of the 2d coordinate plane. If you use 3d coordinates on the sphere, your center of gravity will appear on the unit sphere and the radius can be used to determine how strongly it favors that particular area.

As an example of where this could add confusion, if you have a huge economy in the US and a huge economy in China, you basically are canceling out the values with the current axes, if you shifted them, it may change the plots dramatically with the purely 2d representation.

djvv3• 12 years ago

Interesting, do you have any ideas how to apply this to finding the average of multiple points on a sphere?

jondon• 12 years ago

This is equivalent to what The Economist did. Average of the cartesian points, extend along the radius to the surface.

Houshalter• 12 years ago

One way would be to search for points on the sphere which have the closest (weighted) average distance to all the points. Or some similar metric like that.

So instead of finding a point in the middle of the Earth, it would find a point that is geographically closest to the US and China.

codezero2• 12 years ago

Maybe project the x/y coordinates onto a unit sphere, then find the center of mass in spherical coordinates, then convert those spherical coordinates to the relevant 2d coordinate system you're using, discarding the radius (but keeping it for the information it provides).

throwaway283719• 12 years ago

This is essentially what The Economist did. It results in the 'center of mass' being in the northern hemisphere, somewhere near Greenland, because the two biggest economies are both in the hemisphere but at opposite longitudes.

Sami_Lehtinen• 12 years ago

I guess this is somehow related, world population by latitude and longitude, which you can use to of course derive center.

http://blog.andersen.im/2013/05/interactive-map-of-world-pop...

xvedejas• 12 years ago

I wonder how the center of economic activity would compare to other measures. For instance:

- Center of all landmass

- Center of all arable land

- Center of population

- Center of population weighted by income

Especially if viewed over time, this would give a better sense of the meaning of this data.

magikroundabout• 12 years ago

It seems surprising that by the outbreak of WW1, the economic centre of gravity was hovering to the west of Greece. In other words, all of the European imperial powers scrambling over Africa and the New World still produced less combined output than Asia, Oceania, the Middle East and eastern parts of Europe and Africa.

pitt1980• 12 years ago

Yeah I found that curious too

VLM• 12 years ago

A lot of discussion about the "right" way to calculate a result, without discussing why.

My guess is something along the lines of an assumption that income is roughly equal to spending so you want "the" place on earth thats ideally suited to building your food processing plant or refrigerated warehouse to minimize total air shipping costs assuming everything will ship by air.

I'm not really sure what meaning this "center" has beyond that unless strange assumptions are made, like income is perfectly proportional to capital market size, or income is perfectly proportional to military power or something.

There is very fast alternative discrete rather than continuous method to calculate "a center" that scales very poorly as resolution increases (which doesn't matter because the input data is junk wrt sig figs and truth) which is just to make a giant mesh network of clusters of a discrete billion bucks at a certain lat/lon or whatever, then add an imaginary center that can move that optimizes itself to a minimum distance from all other existing points. You'll get into huge arguments about high enough res and metastability and rounding errors and local maxima/minima but you can ignore all that, given that as an engineering estimate the input data is junk, you just figure the total distance for each points at all whole degree intersections (88 W 43 N aka Chicago-ish, next 89 W 43 N, then 90 W 43 N ... ) so you figure 360*180 (actually more like 178 than 180, and a +2 for polar reasons) and then sort the 64000 or so results and pick the lowest.

Using the discrete method, if you figure there's 64K (16 bits) degree intersections on the globe and maybe 1024 (10 bits) or so clusters of a billion bucks, that is maybe 26 bits worth of distance calcs and additions, figure 3 bits per decimal digit for "less than 10 digits of operation" and we have multicore processors that run about that many ops per second (if you have the memory and IO bandwidth LOL, which you won't), that followed by a very modest sort, so this is quite tractable and has a resolution probably higher than the sig figs in the input data you're feeding it. No, it doesn't scale well to a higher resolution, and thats OK because the input data doesn't warrant it.

The whole topic smells of a really bad dotcom "brain twister" interview question for a CRUD app designer or CSS jockey. Back when that was how it was decided who was a good or bad one based on solving riddles and stuff.

Finding the World's Economic Center of Gravity

Related Stories