Detecting ASICs and other network hashrate variables

[organofcorti]

This is a follow up post to Price drives Difficulty and Forecasting the network hashrate. I describe a model that can be used to detect abnormal changes in the the network.

http://organofcorti.blogspot.com/2012/11/103-canaries-coal-mines-and-black-swans.html

2. Conclusions and discussion
Is model C detecting sudden changes, or is the error due to the lack of a price lag variable? I think the model does both - sudden changes away from the normal linear relationship exceed the 95% confidence interval because of the lack of the price lag variable.

Model C is the proverbial canary. Calculate it each week and find a week that log(H) - model C (log(H, lag1), log(H, lag3),log(P)) > 9.3%, then there's a good chance ASICs are being added. If the converse is true,  some other variable is slowing the response of the network hashrate to the exchange rate. Happy hunting!

• Model C: log(H) ~ 3.96 + 0.52442 log(H, lag = 1) + 0.33676 log(H, lag = 3)  + 0.14132 log(P) +/- 9.3%
• If  log(H) - model C (log(H, lag1), log(H, lag3),log(P)) > 9.3%, the network hashrate is increasing more quickly than indicated by the price (for example ASICs have been switched on).
• If  log(H) - model C (log(H, lag1), log(H, lag3),log(P)) < 9.3%, the network hashrate is increasing more slowly than indicated by the price (for example an increase in D that makes mining unprofitable for many miners.

[bitboyben]

I was thinking about these calculations.

Given BTC's price you can predict the hash rate such that you know if it is a good time to purchase more hardware or not almost in real time.

That would make an awesome online calculator/ alert.
Especially if we get to next day shipping from our asic suppliers.

[organofcorti]

I was thinking about these calculations.

Given BTC's price you can predict the hash rate such that you know if it is a good time to purchase more hardware or not almost in real time.

That would make an awesome online calculator/ alert.
Especially if we get to next day shipping from our asic suppliers.

This one is less about prediction and more about looking for current anomalies in average weekly hashrate. I'll post the long term prediction results soonish. That might be helpful to know going forward.

EDIT I'll do the long term predictions next week and provide the necessary calculations that someone can use to create a calculator.

[bitboyben]

Any abnormal readings?

[organofcorti]

Any abnormal readings?

I'll only do it weekly - I think I'll just update every Sunday when I do the Weekly pool statistics posts.

I'll post here when I start.

That said, I have provided the models and the knowledge of how to use them, so anyone interested can give it a try.

[organofcorti]

The Canary model passes it's first test - the block reward halving was successfully detected! The model predicted a hashrate well higher than it's 95% confidence interval for the percentage error. The network hashrate is probably lower due to the block reward halving, and is thus a variable that is not accounted for by the model. Hopefully other unexpected variables can be detected the same way.


http://organofcorti.blogspot.com/2012/12/weekly-network-forecast-10th-december.html