btcvol
btcvol copied to clipboard
Use multiple imputation to make other currency volatilities fully comparable
Data for fiat currencies is not available for weekends and holidays. Since Bitcoin exchanges operate 365 days/year, computed volatilities are not strictly comparable. We should use a statistical technique called multiple imputation to fill in the missing data for the fiat currencies to do proper comparable volatility estimates.
I'd take a crack at this one. Any libraries you've come across to help with this?
Haven't seen anything for Go. A statistician friend pointed to Amelia for R as a package that would do the job.