Welcome to the Bitcoin Bubble Index, an index to measure the Log-Periodic Power Law Oscillations in Bitcoin
As an example, the chart below shows a backtest simulation of a 90 days forward window for Drawdown 1 to see whether the median numbers of days from the green start line to critical time for a possible burst comes close to matching the actual date of the price crash (orange line).
The corresponding number of "Trading days to the peak" from the simulations can be summarized in a frequency diagram below:
Along with a frequency diagram for the peak price levels obtained from the simulations:
As can be seen in this simulation for Drawdown 1, the median an mode estimates for the 'number of days till the expected crash' and the 'expected peak price level' at that time can be calculated. In this particular simulation, the estimated median critical time (Tc) till crash was 45 days (~10 days early copared to the actual date) with a median equivalent peak price of $1085 (about ~14% below the actual peak price of $1267).
Backtest simulations were runs for 30 day, 60 day and 90 day forward time estimates for all four drawdowns and the following results were generated:
The results suggest that using a 60 day forward window would provide a relatively good balance between forecasting the Critical Time period (Tc) vs. forecasting the expected peak price at that time.
Now the results!
LPPL model using a 60 day forward estimate
The curve plot below corresponds to the optimal parameters found during the global metaheuristic algorithmic search.
On December 20th, 23rd and 29th the Bitcoin Bubble Index of BTC/EUR 52 day, 104 day and 153 day incides peaked respectively.
This means that the derivative reached 0 and indicates that the phase change probability would be highest in the next 21 days.
On January 7th we predicted that we will see a phase change in 21 days from the point of those local peaks.
In summary the greatest probability for a phase change would happen between January 9th and 18th.
At its peak, the BTC/EUR exchang ratw was around €16,310 on December 17, 2017. When the 52 day, 104 day and 153 day indices peaked at the exchange rate closed at €15,800, €14,732 and €13,892.
Looking at the exchange rate data after January 7th, the most noticeable movement was the large draw-down which started on January 6 and still continues in February. If we look at the closing rate on January 9th: €12,025 and compare it to €7,030, the price at its low on February 2nd, we see a draw-down of 44.34%.
Based on this draw-down we would conclude that the prediction on January 7th, based solely on the Bitcoin Bubble Index's quantitative output, was accurate and it remans to be seen what will happen to the BTC/EUR exchange rate, now that the longer term indices have peaked. Our personal opinion is that the rate will continue its downward progression until it reaches significant support, maybe at €6,000 or €5,000 level.