2
$\begingroup$

In the dimensional regularization scheme, four-dimensional integrals are analytically continued from their $d$-dimensional counterparts, i.e.,

$$\int d^4 x\, f(x) \longrightarrow d^d x\, f(x)\,, \tag{1}$$

with, e.g., $d := 4 - \varepsilon$.

My question is simple: can the measure be expanded in $\varepsilon$? I doubt it, but if yes I guess it should take the form

$$d^d x \sim d^4 x - \varepsilon A\, d^4 x + \ldots\,, \tag{2}$$

where $A$ is to be determined.

$\endgroup$

1 Answer 1

1
$\begingroup$
  1. In practice dimensional regularization (DR) is not done by expansion of the measure factor $$d^dx$$ (whatever that is mathematically supposed to mean).

  2. Instead one typically considers the exact analytic expression for the full integral $$I(d)~=~\int\! d^dx~f(x,d)$$ [of the specific class of integrands $f(x,d)$ that one is interested in] as a function of the integer dimension $d\in\mathbb{N}_0$.

    In favorable cases, there exists a natural analytic continuation of $I(d)$ to (a region of) the complex $d$-plane, which then serves as a definition for the DR scheme.

$\endgroup$
1
  • $\begingroup$ In some situations (e.g. the $\varepsilon$-expansion), higher powers of $\varepsilon$ might be needed. Some integrals are too hard to be calculated in arbitrary $d$, so I am looking for an alternative method to relate those higher powers to $4d$ integrals. I should say that I don’t think that it is possible. $\endgroup$
    – Pxx
    Commented Nov 2, 2023 at 9:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.