Topic: Monthly average USD/bitcoin price & trend - page 26. (Read 118254 times)

I believe that bitcoin speculation is way ahead of the underlying bitcoin economy. What will radically change in 2-4 years accordingly, is mainstream perceptions and expectations.


The family of logistic curves I made, and the log growth chart presented by rpietila, probably do not change much if we leave off the last six months of data. The addition of China, and later other countries yield a higher maximum bitcoin price, extending the exponential phase of the logistic curve, not necessarily changing its slope.

Great graphs, rpietila, molecular, and SlipperySlope. I don't know how I missed this thread.

SlipperySlope's graph adds an interesting angle: that we will see mainstream adoption get into full-swing in around 2-4 years. Almost as fun as considering the price growth is considering where Bitcoin will be qualitatively at each order of magnitude.

Mainstream adoption of Bitcoin suggests a radically changed world...in 2-4 years. (Holy shit!)

Also, I have a question for everyone to consider. Suppose Bitcoin never became popular in China. Suppose all investment and adoption stayed limited to the US, Europe, and wherever else it had significant uptake through 2012. We'd still be using this same exponential chart, correct?

If so, wouldn't that mean the addition of China (and later other countries) will have to add an extra doubling or so? It wouldn't be much in the grand scheme of things since it only takes a few months to double anyway, but it might fast-forward the graph by a couple of months, and more or less abruptly starting recently, given the dramatic upsurge in action from China. I know this distracts greatly from the purity of the math here, but then again those additions aren't nothing.

Totally. Did you see this post today? http://www.reddit.com/r/Bitcoin/comments/1qetkx/why_are_so_many_people_here_confident_that/

Sure, since I am not the one doing the charts anyway

Currently it seems to be postponed until May 2014...

No seriously, the value of this analysis is to give you ideas on when we are ahead or behind the trend. Objectively, not based on handdrawings. For example, my decision to time the selling last September was stupid, since the trendline was so much behind that any downside was very unlikely. As I did not have imminent need for the money, I better had waited and sold for example now, as we are ahead the trend (and at triple the price!).

Could we have quarterly vertical lines? It's difficult to pinpoint the exact moment we "will" reach $1000 .

New datapoints used, see OP!

Ah, it's nice to see the growth of BTC. Got to love them!

The reason is merely the convenience of the data series that I downloaded from bitcoincharts.com.

I would guess that changing the starting point would not change the conclusion much - that if indeed bitcoin prices can be modeled according to an innovation adoption logistic function, then the log growth of the past three years may continue for two to four more years before rapidly slowing down. My analysis, if correct, does not say much about what the maximum price might be, but it does say to hurry up and invest now if one's tactic is buy and hold.

So on these graphs, the right side represents:

1×10^x where x is the number.

So:

6=$1,000,000
5=$100,000
4=$10,000
3=$1,000
2=$100
1=$10
0=$1
-1=$.10

Nice graphs, thank you. But is there a reason you selected Mt.Gox opening as the starting point and reject the notion that bitcoin was trading OTC for 1.5 years at an essentially flat rate?

Google Docs. I created a shared spreadsheet that you should be able to copy and modify. The calculations are adapted from the Wikipedia article on the Logistic Function. The math formulas are my own and any errors therein are mine too.

https://docs.google.com/spreadsheet/ccc?key=0ArD8rjI3DD1WdGhDN3FBWFptTlZTREN0cFkxZ3JHTnc

Thanks for your interest.

What program do you use for those graphs?

Here is my hand-fit logistic function chart assuming a maximum bitcoin price of $40000. The Y axis is the Log10 of the price. Observe that according to this model the price continues to increase exponentially through 2015.

Great job, thanks for the charts.
Would it be possible to post something with more conservative numbers, perhaps a max of 40,000.

https://docs.google.com/spreadsheet/ccc?key=0ArD8rjI3DD1WdGhDN3FBWFptTlZTREN0cFkxZ3JHTnc

This spreadsheet is my attempt to fit a logistic function to
the bitcoin price history and thus predict its future.
See Wikipedia http://en.wikipedia.org/wiki/Logistic_function

Because the initial phase of the logistic curve is exponential, I
found that a wide variety of maximum bitcoin prices could
reasonably match the price history. What is interesting is that
the resulting family of curves revealed that the very rapid
growth of bitcoin prices will not last beyond a few more
years.

I configured the calculations to accept a guess as to what the
maximum bitcoin price in $US will be. I tried four values from
$40000 to $5000000. For each maximum bitcoin price, I
manually adjusted the starting X value and adoption period so
that the logistic curve passed through the July 7 2010 price and
also passed through the November 11, 2013 price.

The resulting adoption periods range from 10.3 years for
the $40000 max price to 13.6 years for the $5000000 max
price. The logistic function becomes linear at the adoption
halfway point and then becomes exponentially decreasing.
In the case of a $1000000 bitcoin price, the exponential
growth stops in 2017.

Note that the logistic function is fit to the price curve and
models the behavior of speculators, not the underlying bitcoin
economy - which I suppose may take 20-30 years to become
fully adopted, e.g. replace precious metals and fiat currencies.
Likewise, even after full adoption, bitcoin prices will continue
to grow as the underlying bitcoin economy grows.

Here is the logistic price curve supposing that bitcoin reaches a maximum price of $1000000.
The Y axis is the Log 10 of the price.

And at the subsequent bottom? 1 luxury apartment less?

Right.

Can someone plot dree's data onto a sigmoid graph to see where we are and what is says we will go to?  I am very curious.

I found an thread here that talks about someone using Excel to do something like that:

http://www.excelforum.com/excel-programming-vba-macros/784650-non-linear-regression-fitting-data-to-a-sigmoidal-psychophysical-curve.html

I would do it but I'm too busy this weekend.

In principle, yes. Although if fiat is totally obliterated, also the "long term fair value" of bitcoin may be higher than $300k.