Fix rescaling of LRC and Expansion

Cecca / role-of-dimensionality

MIT License

3 stars 0 forks source link

Closed Cecca closed 3 years ago

Cecca commented 3 years ago

We should use log(Br2 /Br1 )/ log(r2/r1) for the rescaling
In the case of Br2/Br1, pay attention to the base of the logarithm we are using now (should be 2, I'm using natural log)

Cecca commented 3 years ago

After rescaling, the distributions look the same under the two different parameterizations

density-rc

density-expansion

Cecca commented 3 years ago

Fixed by 7ba4e53280db871284dbb48bc08a08f63d0293ee