Review Oscillation Boundary Condition?

[ScatterPlot1]

Hi there

I noticed that your oscillation boundary condition in the confidence indicator construction is set at

O = (w / (2 * np.pi)) * np.log((tc - t1) / (tc - t2)) > 2.5

However, literature (e.g. Sornette et al. (2015), "Real-time prediction and post-mortem analysis of the Shanghai 2015 stock market bubble and crash", p.5) would suggest that

O = (w / 2) * np.log((tc - t1) / (t2 - t1)) > 2.5

I am not sure whether 2.5 may be an applicable value for both definitions, but the division by pi and different log-term definition makes me suspicious of the definition.

[Joshwani]

i believe it is typo in their paper.