The Log Chart Myth

[N12]

Does this really like like a healthy, historically normal correction to you?

[NothinG]

I don't see why everyone has their panties in a wad.

[N12]

I don't see why everyone has their panties in a wad.

Because there is a lot at stake.

Will the log chartists please stand up? Show me evidence this is a normal correction and not an exceptionally deep and long one!

[Elwar]

This:

[fcmatt]

This:

Hey Elwar, you did not start your chart at 0.000001. Why is that?
Are you trying to show some type of bias here that others who post a chart starting at .000001 do not
seem to have? ;-)

[Jixtreme]

*standing*

Math major here, vouching for log chart.

[Elwar]

This:

Hey Elwar, you did not start your chart at 0.000001. Why is that?
Are you trying to show some type of bias here that others who post a chart starting at .000001 do not
seem to have? ;-)

The random chart I found on the Internet did not start at .000001

I just wanted to feel included in the chart posting.

[ElectricMucus]

Log chartist reporting.

What's up?  

srsly what's your problem, just thing of it this way: Blow up the chart to 2030 or so and draw a tanh() funtion over it.
It only behaves linearly (or exponential on a linear chart) because we are looking at a small fraction of it.


here did that for ya:

[greyhawk]

I scrolled to the end of that huge picture in hope there was a funny ending to it.


There wasn't.   

[ElectricMucus]

I scrolled to the end of that huge picture in hope there was a funny ending to it.


There wasn't.   

k fixed

Sorry not very creative 

[d'aniel]

Log chartist reporting.

What's up?  

srsly what's your problem, just thing of it this way: Blow up the chart to 2030 or so and draw a tanh() funtion over it.
It only behaves linearly (or exponential on a linear chart) because we are looking at a small fraction of it.

Are you modelling the diffusion of Bitcoin with the logistic differential equation?  Is that why you used tanh?

[ElectricMucus]

Log chartist reporting.

What's up?  

srsly what's your problem, just thing of it this way: Blow up the chart to 2030 or so and draw a tanh() funtion over it.
It only behaves linearly (or exponential on a linear chart) because we are looking at a small fraction of it.

Are you modelling the diffusion of Bitcoin with the logistic differential equation?  Is that why you used tanh?

I'm sorry I have no idea what you are talking about  

But you might be on to something here 

[d'aniel]

Log chartist reporting.

What's up?  

srsly what's your problem, just thing of it this way: Blow up the chart to 2030 or so and draw a tanh() funtion over it.
It only behaves linearly (or exponential on a linear chart) because we are looking at a small fraction of it.

Are you modelling the diffusion of Bitcoin with the logistic differential equation?  Is that why you used tanh?

I'm sorry I have no idea what you are talking about  

But you might be on to something here 

Oh.  If you're interested: http://en.wikipedia.org/wiki/Diffusion_of_innovations.  Notice the graph at the top of the page, and the "logistic function" link underneath.

[ElectricMucus]

Log chartist reporting.

What's up?  

srsly what's your problem, just thing of it this way: Blow up the chart to 2030 or so and draw a tanh() funtion over it.
It only behaves linearly (or exponential on a linear chart) because we are looking at a small fraction of it.

Are you modelling the diffusion of Bitcoin with the logistic differential equation?  Is that why you used tanh?

I'm sorry I have no idea what you are talking about  

But you might be on to something here 

Oh.  If you're interested: http://en.wikipedia.org/wiki/Diffusion_of_innovations.  Notice the graph at the top of the page, and the "logistic function" link underneath.

thanks, I've already looked it up, pretty interesting.

But we have a problem: Scaling the function to the actual data would be quite a demanding task, and would require skillfull programming instead of naive curve drawing by hand 

[wee baby seamus]

are you guys just drawing lines with paint, or are you actually fitting to some curve or something? nothing's properly labeled so I can't tell, but it seems like you're just making this shit up

[Synaptic]

are you guys just drawing lines with paint, or are you actually fitting to some curve or something? nothing's properly labeled so I can't tell, but it seems like you're just making this shit up

Welcome to Bitcoin.

Please purchase your tokens now.

[d'aniel]

Log chartist reporting.

What's up?  

srsly what's your problem, just thing of it this way: Blow up the chart to 2030 or so and draw a tanh() funtion over it.
It only behaves linearly (or exponential on a linear chart) because we are looking at a small fraction of it.

Are you modelling the diffusion of Bitcoin with the logistic differential equation?  Is that why you used tanh?

I'm sorry I have no idea what you are talking about  

But you might be on to something here 

Oh.  If you're interested: http://en.wikipedia.org/wiki/Diffusion_of_innovations.  Notice the graph at the top of the page, and the "logistic function" link underneath.

thanks, I've already looked it up, pretty interesting.

But we have a problem: Scaling the function to the actual data would be quite a demanding task, and would require skillfull programming instead of naive curve drawing by hand 

lol

Plus, I think you'd want to use the linear scaling when fitting the logistic function.  Would be a pretty bad fit, then

[d'aniel]

are you guys just drawing lines with paint, or are you actually fitting to some curve or something? nothing's properly labeled so I can't tell, but it seems like you're just making this shit up

Mostly just having fun with some math.

I wouldn't place much confidence in the predictive power of this model.  Too many significant hurdles yet to be overcome to say it's just a simple matter of spreading the word.

[Minsc]

[Surawit]

You don't need to use a log chart. Simply view the standard price chart

http://bitcoincharts.com/charts/mtgoxUSD#rg60zvztgSzm1g10zm2g25 > set axis to 6-months

Now, you simply apply a Milton-Keynesian correction transform to the market data, like so:

https://i.imgur.com/qqtEf.png

Using this metric it is clear the value of bitcoin has stabilised on the horizontal. BUY BUY BUY