Post
Topic
Board Securities
Re: Lab Rat Data Processing, LLC (LabRatMining) Official Announcement
by
mmmerlin
on 12/11/2013, 18:24:38 UTC
In formulating a model to look at bonds and give individuals a sense of how they may want to position a trade, I noticed an interesting niggling error.

The last "official" excel sheet that LR has posted noted 55472 bonds held.  In the most recent dividend distribution, BTC31.01295233 was distributed, at a per bond rate of BTC.000557995.  I calculate that to imply that held bonds total 55579.  So, 127 bonds in the overall scheme of things isn't so many, but it still amounts to a 2.2% discrepancy from the "official" total when compared to what was distributed.  Not sure how that works out.

I haven't fully mined the bitfunder data, so I don't have numbers there, but for some preliminary numbers, I noted that the week ending 11/9 (our "business week" basically ends on Saturdays, since that's the distribution day), there were 4 transactions that amounted to 183 bonds in open interest.  Total BTC volume was 9.9635, with an average price, therefore, being .0544453552.  That would imply an average yield for that week to be 1.0249%.

This week (week ending 11/16) is not yet done, obviously, but thusfar, volume already surpasses the previous week (up to 5 transactions), with open interest also exceeding, at 209.  Price action, however, is down to .0507093301, down nearly 6.9%.  The downtick on the bond pricing, however, should be taken with a grain of salt since we have many trading days ahead of us.

Anyway, as more data becomes available, I'll see about making the reporting more meaningful.

PM me any thoughts or comments on metrics that might be useful, and I'll see about building some spreadsheets.

Numbers taken from these very low liquidity markets should be expressed with huge error bars IMHO. This is very useful info, and I thank you for collecting it and compiling it, but large error bars must be used to stop people extrapolating from very small data sets. (Many people are just really stupid I'm afraid, and will read more into these metrics than anyone ever should.)

Thanks again though, I think this is an important job.