CEA45 Archived on Twitter: "To better visualize observed data, we also continually update a curve-fitting exercise to summarize COVID-19's observed trajectory. Particularly with irregular data, curve fitting can improve data visualization. As shown, IHME's mortality curves have matched the data fairly well.… https://t.co/8f3wGOxll0"

Country	Code	For customers of
United States	40404	(any)
Canada	21212	(any)
United Kingdom	86444	Vodafone, Orange, 3, O2
Brazil	40404	Nextel, TIM
Haiti	40404	Digicel, Voila
Ireland	51210	Vodafone, O2
India	53000	Bharti Airtel, Videocon, Reliance
Indonesia	89887	AXIS, 3, Telkomsel, Indosat, XL Axiata
Italy	4880804	Wind
3424486444	Vodafone
» See SMS short codes for other countries

1. Jack Ross‏ @jwross55 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  What use is the cubic fit plot? When you inform policy makers, what do you say the cubic fit plot tells you?
  
  9 replies . 1 retweet 417 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Leanne‏ @leanneotero 5 May 2020
  Replying to @WhiteHouseCEA45
  
  pic.twitter.com/f5tdioE5zZ
  
  7 replies . 61 retweets 328 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Brian Bohman‏ @bjbohman 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  Brian Bohman Retweeted Nate Silver
  
  Looks more like MS Paint than MS Excel to me but I think someone may owe @NateSilver538 $538 on this one...https://twitter.com/NateSilver538/status/1257476755574718470?s=20 …
  
  Brian Bohman added,
  
  Nate SilverVerified account @NateSilver538
  
  I would bet $538 that the White House's "cubic model" is literally just an MS-EXCEL trendline with a third-degree (cubic) polynomial. https://twitter.com/jdawsey1/status/1257395401461727232 …
  
  Show this thread
  
  1 reply . 11 retweets 211 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Tom‏ @TomTomClub510 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  Tom Retweeted David Leavitt
  
  https://twitter.com/david_leavitt/status/1257561269370060801?s=21 …
  
  Tom added,
  
  David LeavittVerified account @David_Leavitt
  
  US #coronavirus deaths 9 weeks ago: 6 deaths 8 weeks ago: 26 deaths 7 weeks ago: 86 deaths 6 weeks ago: 541 deaths 5 weeks ago: 3003 deaths 4 weeks ago: 10,986 deaths 3 weeks ago: 23,649 deaths 2 weeks ago: 42,295 deaths 1 week ago: 56,255 deaths Right now: 70,000 deaths
  
  4 replies . 15 retweets 47 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Colin Camerer‏ @CFCamerer 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  This is so amateurish. Kernel smooth for irregular dataviz. Original IHME still looks best overall, which suggests revisions were just overfitting
  
  3 replies . 3 retweets 81 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. John Stone‏ @jnzst 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  Economists shouldn't talk about epidemiology, example 49383
  
  8 replies . 6 retweets 139 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. David Anderson‏ @bjdickmayhew 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  So, with this "cubic fit" it fundamentally assumes that no one has new infections since roughly April 27 as death is a multi-week lagging function of exposure and infections. Stay away from the EXCEL functions.
  
  3 replies . 10 retweets 171 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Brian Fox‏ @fox_bfox 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  You did not just do this. OMG, this wouldn't pass an introductory data analysis class.
  
  2 replies . 8 retweets 218 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Brian Bohman‏ @bjbohman 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  Brian Bohman Retweeted Sean J. Taylor
  
  The key logical fallacy laid bare here is that a "curve fit" has inherent use value for "forecasting". Great visual from @seanjtaylor showing the disconnect between interpolation (i.e., curve fitting) and extrapolating (i.e., forecasting)https://twitter.com/seanjtaylor/status/1123278396308353024?s=20 …
  
  Brian Bohman added,
  
  Sean J. Taylor @seanjtaylor
  
  Your first attempt at fitting a curve might be a polynomial. You can see that as you add terms you can *interpolate* better but the *extrapolation* is very poor. This is because the local trend it is estimating is usually very wrong. pic.twitter.com/LIZz6vjSeR
  
  Show this thread
  
  5 replies . 83 retweets 496 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo
1. Scott Delhommer‏ @s_delhommer 5 May 2020
  Replying to @WhiteHouseCEA45 @WhiteHouseCEA
  
  literally all economists when seeing this. please.pic.twitter.com/2tTC6T5OjP
  
  8 replies . 18 retweets 630 likes
  Thanks. Twitter will use this info to make your timeline better. Undo
  
  Undo