# Last edited on 2014-05-12 00:18:19 by stolfilocal

[b]Chinese Slumber Method (non-)prediction for Monday May 12[/b]

There is no prediction for Monday.  In fact there will be no more Chinese Slumber Method (CSM) predictions. [ Waiting for the hoorays to subside. ] A short report about it is in the works, and the conclusions are disappointing: while CSM is more accurate than the banal predictor "tomorrow is like today" (TLT) on about 60% of the days, on the other 40% is so much worse that it loses to TLT, by most other measures.

[b]Checking the previous prediction[/b]

Prediction was posted on: Sunday 2014-05-11, 02:04 UTC 
Prediction was valid for: Sunday 2014-05-11, 19:00--19:59 UTC (~17 hours later)

Foiled again again [i]again[/i] by the treacherous Chinese:

Huobi's predicted price: 2823 CNY
Huobi's actual price (L+H)/2: 2687 CNY
Error: 136 CNY (~22 USD)

Bitstamp's predicted price: 456 USD
Bitstamp's actual price (L+H)/2: 438 USD
Error: 18 USD
 
[url=http://www.ic.unicamp.br/~stolfi/temp/2014-05-11-prediction-check-HUBI-annotated.png][img]http://www.ic.unicamp.br/~stolfi/temp/2014-05-11-prediction-check-HUBI-annotated.jpg[/img][/url]

[url=http://www.ic.unicamp.br/~stolfi/temp/2014-05-11-prediction-check-BSTP-annotated.png][img]http://www.ic.unicamp.br/~stolfi/temp/2014-05-11-prediction-check-BSTP-annotated.jpg[/img][/url]

[ [url=http://www.ic.unicamp.br/~stolfi/temp/2014-05-11-prediction-check-legend.txt]Plot legend[/url] ]
 
[b]NOTE[/b]: 
  "Erasmus Darwin had a theory that once in a while one should perform a
  damn-fool experiment. It almost always fails, but when it does come
  off is terrific. Darwin played the trombone to his tulips. The result
  of this particular experiment was negative."
    -- J.E.Littlewood