Daily volumes of BTC trade to/from USD and other national currencies (in kBTC):

             !    Thu !    Fri !    Sat !   Sun !    Mon !    Tue !    Wed !    Thu !
  EXCHANGE   !  03/13 !  03/14 !  03/15 ! 03/16 !  03/17 !  03/18 !  03/19 !  03/20 ! Currencies considered

  Bitstamp   |   6.73 |   8.68 |   2.20 |  3.30 |   9.17 |  14.90 |   7.08 |  10.90 | USD
  BitFinEx   |   3.31 |   5.53 |   1.74 |  1.90 |   3.76 |   7.74 |   3.44 |   8.10 | USD
  BTC-e      |   7.43 |   3.76 |   2.55 |  2.62 |   3.90 |   7.91 |   5.48 |   7.85 | USD,EUR,RUR
  Kraken     |   0.40 |   0.48 |   0.34 |  0.15 |   0.48 |   0.85 |   0.57 |   0.74 | EUR
  CampBX     |   0.03 |   0.06 |   0.03 |  0.02 |   0.07 |   0.32 |   0.06 |   0.72 | USD
  CaVirtEx   |   0.13 |   0.11 |   0.06 |  0.06 |   0.10 |   0.10 |   0.13 |   0.40 | CAD
  Bitcoin.DE |   0.35 |   0.20 |   0.12 |  0.12 |   0.40 |   0.42 |   0.35 |   0.38 | EUR

  SUBTOTAL   |  18.38 |  18.82 |   7.04 |  8.17 |  17.88 |  32.24 |  17.11 |   0.00 |

  OKCoin     |  60.70 |  97.80 |  55.40 | 40.00 |  59.00 |  82.70 |  95.00 |  58.30 | CNY
  Huobi      |  65.40 |  78.40 |  36.50 | 27.00 |  63.80 |  84.10 |  62.40 |  48.10 | CNY
  BTC-China  |   2.89 |   2.97 |   1.13 |  1.51 |   2.74 |   3.70 |   3.79 |   2.85 | CNY
  Bter       |   0.26 |   0.19 |   0.35 |  0.28 |   0.34 |   0.27 |   0.35 |   0.36 | CNY

  SUBTOTAL   | 129.25 | 179.36 |  93.38 | 68.79 | 125.88 | 170.77 | 161.54 |   0.00 |

  TOTAL      | 147.63 | 198.18 | 100.42 | 76.96 | 143.76 | 203.01 | 178.65 |   0.00 |

All numbers were collected by hand from the site http://bitcoinwisdom.com. Beware of possible errors.

For each exchange, the numbers include only the trade volume to/from the currencies listed in the rightmost column. Trade between BTC and other cryptocoins, such as LiteCoin, is NOT included.

Dates on the header line are UTC. Specifically, "01/15" means "from 01/15 00:00:00 UTC to 01/15 23:59:59 UTC". (Beware that the same period may span part of day 01/14 or part of day 01/16 in your local time zone.)

NOTE: There are at least five additional large exchanges in China that are not on this table because they are not covered by Bitcoinwisdom: FxBTC, ChBTC, BTCTrade, BTC100, and BTC38.  Together they may have as much volume as Huobi or OKCoin.

Chinese Slumber Method prediction for Friday March 21

Prediction valid for: Friday 2014-03-21, 19:00--19:59 UTC (not before, not after)
Huobi's predicted price: 3630 CNY.
Bitstamp's predicted price: 591 USD.



The red and green strokes are actual hourly prices (Huobi on the left, Bitstamp on the right).  The magenta square at right is the above prediction.  The blue square next to it is the last prediction (see below). The light blue-gray squares are the older predictions.   The blue dots are the Slumber Points, the mean prices ((L+H)/2) in the interval 19:00--19:59 UTC every day. On the Huobi plot, the size of the Albertosaurus under each dot indicates the night-time activity at Huobi. Specifically, the area of the reptile is proportional S = V_h/V_d, where V_h is Huobi's mean hourly volume in the three hour period 18:00--20:59 UTC (02:00am--04:59am China time), and V_d is Huobi's daily volume 00:00--23:59 UTC on the same date.  The largest Albertosauri correspond to the S = 0.010 or greater.   The size of each blue dot (on both plots) is proportional to its reliability weight, computed from Huobi's ratio S.  The brown lines on Huobi's plot are trends fitted a posteriori to the Slumber Points. The orange line is the trend assumed for the Huobi prediction above.

Specifically, the trend used is a straight line A + B*(d-d0) where d is the day of the month, d0 = 16, A = 3818.88, B = -37.70, fitted by weighted least squares to the points Mar/16--20.  (The fit was not very good, so it was tempting to assume a trend break between Mar/18 and Mar/19.  But let's wait for the next data point before deciding that.)

The Bitstamp prediction, as usual, is the Huobi prediction divided by the currency conversion factor R, which on Mar/19 seems to have jumped from ~6.06 to ~6.14 CNY/USD.

Checking the previous prediction

Prediction was posted on: Wednesday 2014-03-19, 22:24 UTC
Prediction was valid for: Thursday 2014-03-20, 19:00--19:59 UTC (~20.5 hours later)

Huobi's predicted price: 3725 CNY.
Huobi's actual price (L+H)/2: 3646 CNY
Error: 79 CNY (~13 USD)

Bitstamp's predicted price: 615 USD.
Bitstamp's actual price (L+H)/2: 591 USD
Error: 24 USD

NOTE:

Isn't this what he said right away?

That is great!  Thanks!!

Would you trust your money to this:  NYBX - New York Bitcoin Exchange (Not to be confused with NYBX - New York Block Exchange, a branch of NYSE closed feb/2013)

(Hey, it has an academic on board!)

My highly technical (academic even) trend analysis (NOT based on Chinese unhealthy sleep habits) says 430 USD by Apr/01, 330 USD by May/01.

These forecasts are totally free and guaranteed, if they fail your payment will be refunded some day (minus postage and handling). 

I  see.  Well, but if those stored don't touch bitcoins, they cannot be counted as bitcoin users, can they?

By that criterion, every store would be a "bitcoin user", because every store accepts money that was obtained by selling BTC.

Er, sorry for the ignorance, but why would one use Bitpay if one owns no bitcoins?

An interesting statistic one could get from the blockchain is the net amount of USD that actually went out those addresses.  Namely, for each transfer out of an address one adds the BTC amount times the USD/BTC rate at that time; and subtracts the same formula for any transfer into the address. 

E.g. if an address received only 2 BTC when the price was 100$ and sent 1 BTC when it was 800$, the balance at that address would be 1 BTC and its net USD outflow would be 600$.

Some addresses will then hold a negative amount.  Indeed, since transfers between addresses add and subtract the same amount, the sum of all such outflows will be negative, equal in magnitude to the sum of all mined coins times the price at the time they were mined.

You mean exchanges, tumblers, Bitpay?  But people who use those probably have at least one private address.  So the number of addresses can only be higher than the number of owners.

0.001 BTC is 0.60 USD.  I doubt that anyone would care about that, not worth one's time to fetch the keys etc. -- unless one has 10,000 such addresses and some automated system to merge or manage them.

So I would exclude accounts below 0.01 BTC (6 USD), and even a fraction of those below 0.1 BTC (60  USD).

"How many systems administrators are needed to replace a light bulb?" "None, the SA will just lock the room and declare it off-limits to everyone."

The point is that the effect of RMB (or USD) inflation on BTC price is well below the noise level.

BTC could be an alternative in countries with, say, 50% inflation per month; but there USD is usually much more attractive than BTC.

Indeed, it is headlines when USD/RMB drops from 6.14 to 6.22 in a month.  I just saw BTC drop from 620 to 590 in a few minutes...

The markets are closed in RMB for common mortals, but some people may have channels, e.g. people who run import/export firms.

In any case, since bitcoins can move freely in and out, arbitrage between Chinese and non-Chinese exchanges is possible even if there is no USD/RMB exchange (I gave a simplistic example some time ago).  Which explains why the price is essentially the same in all exchanges in the world, with changes propagating almost instantaneously. Contrast that with the last days of MtGOX, when that market was really isolated, in BTC as well as in dollars.

While the raw percentage above may not mean much, volume in China appears to be correlated with volatility everywhere.  The lull during the Chinese New Year week is one evidence of that, and one can see on the Western exchanges the effect of the Chinese traders going to bed around 18:00 UTC.  Therefore, the fact that the percentage rose from 60% to 90% after MtGOX's collapse is quite significant IMHO.

Higher volume, for whatever cause, seems to make the price more stable on smaller time scales (cehck the 1-minute charts).  It may also mean more liquidity, and this presumably means more weight when dfining the price (like a rotweiler and a poodle on the same leash: the poodle may sometimes lead, but most of the time it will have to go where the other wants to go).

There may be a lot of off-exchange trade, but how do those traders fix the price, if not by the open market? The BIT fund, AFAIK, computes the nominal value of its shares daily from the BTC/USD price at various exchanges.

EDIT: typos

Total trade volume today (Wed Mar/19 00:00--23:59 UTC), on the exchanges that I monitor, was ~179 kBTC.  That is rather on the low side for the last few weeks; it is 12% less than yesterday's and 11% less than last Wednesday (Mar/12).

Volume outside China fell 47% from yesterday's  (32 to 17 kBTC). Bitstamp (7.08 kBTC), BTC-e (5.48) and Bitfinex (3.44) kept their ranks, but Bitfinex fell further behind BTC-e.  Yesterday CampBX had a sudden surge from 0.07 to 0.32 kBTC, but today it returned to its usual range with 0.06 kBTC.

Volume in the four Chinese exchanges that I monitor a dropped only 5.4% (from 171 to 162 kBTC). OKCoin again passed Huobi (with 59% of my Chinese total against 39%).

It should be noted that there are several large exchanges in China that I do not monitor because they are not listed by Bitcoinwisdom.  They may account for 1/3 or more of the volume in that country.  My volume table also omits many exchanges outside China, but I believe they have very small volume. In spite of that, China's slice of my total volume is 90%.

Daily volumes of BTC trade to/from USD and other national currencies (in kBTC):

             !    Wed !    Thu !    Fri !    Sat !   Sun !    Mon !    Tue !    Wed !                     
  EXCHANGE   !  03/12 !  03/13 !  03/14 !  03/15 ! 03/16 !  03/17 !  03/18 !  03/19 ! Currencies considered

  Bitstamp   |  12.70 |   6.73 |   8.68 |   2.20 |  3.30 |   9.17 |  14.90 |   7.08 | USD                 
  BTC-e      |   7.96 |   7.43 |   3.76 |   2.55 |  2.62 |   3.90 |   7.91 |   5.48 | USD,EUR,RUR         
  BitFinEx   |   7.29 |   3.31 |   5.53 |   1.74 |  1.90 |   3.76 |   7.74 |   3.44 | USD                 
  Kraken     |   0.56 |   0.40 |   0.48 |   0.34 |  0.15 |   0.48 |   0.85 |   0.57 | EUR                 
  Bitcoin.DE |   0.37 |   0.35 |   0.20 |   0.12 |  0.12 |   0.40 |   0.42 |   0.35 | EUR                 
  CaVirtEx   |   0.13 |   0.13 |   0.11 |   0.06 |  0.06 |   0.10 |   0.10 |   0.13 | CAD                 
  CampBX     |   0.04 |   0.03 |   0.06 |   0.03 |  0.02 |   0.07 |   0.32 |   0.06 | USD                 

  SUBTOTAL   |  29.05 |  18.38 |  18.82 |   7.04 |  8.17 |  17.88 |  32.24 |  17.11 |                     

  Huobi      |  77.90 |  65.40 |  78.40 |  36.50 | 27.00 |  63.80 |  84.10 |  62.40 | CNY                 
  OKCoin     |  90.60 |  60.70 |  97.80 |  55.40 | 40.00 |  59.00 |  82.70 |  95.00 | CNY                 
  BTC-China  |   3.16 |   2.89 |   2.97 |   1.13 |  1.51 |   2.74 |   3.70 |   3.79 | CNY                 
  Bter       |   0.49 |   0.26 |   0.19 |   0.35 |  0.28 |   0.34 |   0.27 |   0.35 | CNY                 

  SUBTOTAL   | 172.15 | 129.25 | 179.36 |  93.38 | 68.79 | 125.88 | 170.77 | 161.54 |                     

  TOTAL      | 201.20 | 147.63 | 198.18 | 100.42 | 76.96 | 143.76 | 203.01 | 178.65 |                     

All numbers were collected by hand from the site http://bitcoinwisdom.com. Beware of possible errors.

For each exchange, the numbers include only the trade volume to/from the currencies listed in the rightmost column. Trade between BTC and other cryptocoins, such as LiteCoin, is NOT included.

Dates on the header line are UTC. Specifically, "01/15" means "from 01/15 00:00:00 UTC to 01/15 23:59:59 UTC". (Beware that the same period may span part of day 01/14 or part of day 01/16 in your local time zone.)

NOTE: There are at least five additional large exchanges in China that are not on this table because they are not covered by Bitcoinwisdom: FxBTC, ChBTC, BTCTrade, BTC100, and BTC38.  Together they may have as much volume as Huobi or OKCoin.

Is this the so-called "blockchain" that you folks talk so much about?  I didn't realize it was so huge...

That WSJ article was posted  here but does not mention liquidity.  Where is that?

Chinese Dinosaur Slumber Method prediction for Thursday Mar/20

Prediction valid for: Thursday 2014-03-20, 19:00--19:59 UTC (not before, not after)
Huobi's predicted price: 3725 CNY.
Bitstamp's predicted price: 615 USD.


The red and green strokes are actual Huobi hourly prices.  The magenta square at right is the above prediction.  The blue square next to it is the last prediction (see below). The light blue-gray squares are the older predictions.   The blue dots are the Slumber Points, the mean Huobi prices ((L+H)/2) in the interval 19:00--19:59 UTC every day. The size of the Albertosaurus under each dot indicates the night-time activity at Huobi. Specifically, the area of the reptile is proportional S = V_h/V_d, where V_h is the mean hourly volume in the three hour period 18:00--20:59 UTC (02:00am--04:59am China time), and V_d is the daily volume 00:00--23:59 UTC on the same date.  The largest Albertosauri correspond to the S = 0.010 or greater.   The size of each blue dot is proportional to its reliability weight, computed from the ratio S.  The brown lines are trends fitted a posteriori to the Slumber Points. The orange line is the trend assumed for the above prediction.

The trend used is a shifted exponential A + B*Q**(d-d0) where d is the day of the month, d0 = 16, A = 3722.25, B = 112.56, Q = 0.400.

The Bitstamp prediction, as usual, is the Huobi prediction divided by the currency conversion factor R, which these days seems to be around 6.06 CNY/USD.

Checking the previous prediction

Prediction was posted on: Wednesday 2014-03-17, 05:29 UTC
Prediction was valid for: Wednesday 2014-03-17, 19:00--19:59 UTC (~13.5 hours later)

Well, I was lucky it seems:

Huobi's predicted price: 3699 CNY.
Huobi's actual price (L+H)/2: 3746 CNY
Error: 47 CNY (~8 USD)

Bitstamp's predicted price: 610 USD.
Bitstamp's actual price (L+H)/2: 609 USD
Error: 1 USD

NOTE: This note intentionally left blank.