# Last edited on 2014-03-09 22:49:47 by stolfilocal

[b]Chinese Boring Even With Dinosaurs Slumber Method prediction for Monday Mar/10[/b]

[b]Prediction valid for:[/b] Monday 2014-03-10, 19:00--19:59 UTC (not before, not after)
[b]Huobi's predicted price:[/b] 3938 CNY.
[b]Bitstamp's predicted price:[/b] 656 USD.

[b]Huobi[/b]
[url=http://www.ic.unicamp.br/~stolfi/temp/2014-03-09-prediction-check-HUBI-annotated.png][img width=600]http://www.ic.unicamp.br/~stolfi/temp/2014-03-09-prediction-check-HUBI-annotated.png[/img][/url]
The red and green strokes are actual Huobi hourly prices.  The current prediction is the rightmost magenta square.  The blue square is the last prediction (see below), and the light blue-gray squares are the previous ones.  The orange and grey dots are the Slumber Points, the mean Huobi prices at 19:00 UTC every day.  Each point is a Glyptodon if the hourly volume V[sub]h[/sub] for 19:00--19:59 UTC is less than 0.005 times the daily volume V[sub]d[/sub] 00:00--23:59 UTC; and is an Albertosaurus otherwise. The grey lines are trends fitted [i]a posteriori[/i] to the Glyptodon Points. The orange line is the trend that was assumed for the prediction. The Germanodactyl (hereby assumed to be an ancestor of [i]Bos primigenius[/i]) indicates that the trend is ascending.

Today again was a Glyptodon, again barely (V[sub]h[/sub]/V[sub]d[/sub] = 0.00487 < 0.005).  This point is obviously not on the same trend as the Mar/04--06 Glyptodons, so it seems reasonable to assume a new trend, the straight line defined by the last two Glyptodons; namely, p(d) = A+B*(d-8), where d is the day of the month, A = 3668 and B = +135.  

[b]Bitstamp[/b]
[url=http://www.ic.unicamp.br/~stolfi/temp/2014-03-09-prediction-check-BSTP-annotated.png][img width=600]http://www.ic.unicamp.br/~stolfi/temp/2014-03-09-prediction-check-BSTP-annotated.png[/img][/url]
The red and green strokes are actual Bitstamp hourly prices.  The dots, dinosaurs, mammals, lines, and squares are [b]Huobi[/b]'s Slumber Points, dinosaurs, mammals, trends, and predictions, scaled by the currency conversion factor R.

The actual price ratio between Huobi and Bitstamp has been has been below 6.00 for most of last week:
[url=http://www.ic.unicamp.br/~stolfi/temp/2014-03-09-slumber-points-HUBI-BSTP-ratio.png][img width=600]http://www.ic.unicamp.br/~stolfi/temp/2014-03-09-slumber-points-HUBI-BSTP-ratio.png[/img][/url]
Therefore it seems necessary to adopt a new value of R in order to map the predicitons from one to the other. Accordingly I am now using R = 6.40 for Feb/07--09, R = 6.00 since Mar/04, and R = 6.12 for all other times before  Mar/04.  This new value of R was used to map the points and trends above, and the new prediction (the magenta square), but not the past predictions.

[b]Checking the previous prediction[/b]

Prediction was posted on: Saturday 2014-03-08, 21:54 UTC 
Prediction was valid for: Sunday 2014-03-09, 19:00--19:59 UTC 

Huobi's predicted price: 3596 CNY
Huobi's actual price (L+H)/2: 3803 CNY
Error: 207 CNY (~35 USD)

Bitstamp's predicted price: 588 USD
Bitstamp's actual price (L+H)/2: 641 USD
Error: 53 USD 

[b]NOTE[/b]: Back to chicken-chasing mode, it seems :(