Possible bug in Higgs mass calculation

[JoseEliel]

In the function m2Lambda_to_vMh2 (defined in line 38 of smpar.py) the Higgs mass formula

    Mh2 = 2 * m2 * (1 - m2 / Lambda * (3 * C['phi'] - 4 * Lambda * C['phiBox'] +
                                        Lambda * C['phiD']))

seems to have a typo. Comparing with the expression from https://arxiv.org/abs/1704.03888, there should be a Lambda^2 term in the factor multiplying the second parenthesis, i.e. it should be

Mh2 = 2 * m2 * (1 - m2 / (Lambda*Lambda) * (3 * C['phi'] - 4 * Lambda * C['phiBox'] + 
                                       Lambda * C['phiD']))

[DavidMStraub]

Oh. I don't see why this would not be a bug, indeed.

Does changing that solve #52?

[JoseEliel]

No, I have just checked. I think they are unrelated issues.

[DavidMStraub]

Waiting for @peterstangl or @jasonaebischerGIT to confirm.

Then it would be great if you can submit a PR (but please use **2, this is not Fortan 😉)

[JoseEliel]

I will. Good point on **2, I'm often writing C++ (luckily not Fortran 😉 ), so that's where that came from.

[jasonaebischerGIT]

... there should be a Lambda^2 term in the factor multiplying the second parenthesis, i.e. it should be

Mh2 = 2 * m2 * (1 - m2 / (Lambda*Lambda) * (3 * C['phi'] - 4 * Lambda * C['phiBox'] + 
                                       Lambda * C['phiD']))

I agree, this formula is correct.

[JoseEliel]

I've made the pull request.

[jasonaebischerGIT]

The check from travis seems to have failed. Not sure what the origin of this is...

[celis]

from the travis logs it seems one needs to reduce the numerical accuracy (places argument) required in the assertAlmostEqual method.

[peterstangl]

This issue should be solved by the new PR #59 and I think PR #58 can be closed.

BTW: Fortran actually does implement the ** exponentiation operator. So this seems to be just an issue with C/C++ 😉
(that's why one can get relatively nice expressions that work in Python from Mathematica's FortranForm)

[jasonaebischerGIT]

Closing this issue, which is resolved by PR #59.