Economics

LN Turnover Variance

It has been said that “All models are wrong but some are useful.” In other words, any model is at best a useful fiction. As there is often no better alternative for innovations than to initially make rough assessments based on vague assumptions, this approach was used to estimate the unknown. The use of average values has resulted in approximate values for the turnover variance that appears to be appropriate.

For experimental purposes, it is assumed that the LN turnover facilitated by Plenny fluctuates randomly. Consequently, this variance is probably affecting the token price. When modeling the randomly expected turnover facilitated in bitcoin by Plenny over the LN, the calculations follow a specific formula. The parameters applied relate to the exchange rate between cryptocurrency (sat and PL2) and fiat currency (e.g. USD).

Plenny operates under the assumption it will achieve 200% of its projected turnover in a best-case scenario. In contrast, it will achieve only 50% in a worstcase scenario while still keeping the most likely system at the 100% turnover target achieved. Simulation tools for game design have been used to justify this supposition.

In the statistical literature there is evidence of various distribution models suitable for modelling uncertainty. If turnover is considered as a number of events, the Poisson distribution could be used. In practice, the beta distribution model was used because its application to test data gave a more flexible fit than other models.

Turnover assumptions are random estimates (i.e. “guestimates”). Those numbers are likely to vary widely when real transactions commence. Using a beta distribution (a standard approach for modeling uncertainty) for the expected range of achieved turnover, the predicted range for the token price is modelled very roughly.

Find more details: Working Paper