2002
DOI: 10.1016/s0920-5632(01)01638-3
|View full text |Cite
|
Sign up to set email alerts
|

Constrained curve fitting

Abstract: We survey techniques for constrained curve fitting, based upon Bayesian statistics, that offer significant advantages over conventional techniques used by lattice field theorists.

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
246
0

Year Published

2004
2004
2023
2023

Publication Types

Select...
7
2

Relationship

0
9

Authors

Journals

citations
Cited by 288 publications
(249 citation statements)
references
References 4 publications
3
246
0
Order By: Relevance
“…[20,21]. In Secton II we showed how a "naturalness prior" obtained using the principle of maximum entropy leads straightforwardly to an augmented χ 2 .…”
Section: Discussionmentioning
confidence: 95%
See 1 more Smart Citation
“…[20,21]. In Secton II we showed how a "naturalness prior" obtained using the principle of maximum entropy leads straightforwardly to an augmented χ 2 .…”
Section: Discussionmentioning
confidence: 95%
“…(24) and (25) we can write this posterior in the form pr(a res |D, M, R) ∝ da marg exp − 1 2 χ 2 aug , where χ 2 aug = a T A aug a − 2b · a + C for the linear case, and A aug , b and C are given in Eqs. (21), (19), and (20) respectively. We now show that marginalization over a marg does not change the result for the expectation value of a res and its variance.…”
Section: Appendix B: Marginalization Over Higher-order Parameters In mentioning
confidence: 99%
“…We fit 2-and 3-point correlators simultaneously using Bayesian methods [42] to constrain fit parameters and determining the covariance between results at different Q 2 values. The fit forms are [31,37] …”
Section: Lattice Qcd Calculationmentioning
confidence: 99%
“…A straightforward fit to a finite sum of exponentials is sometimes possible, usually when high statistics are available. Constrained fitting [1] and the Maximum Entropy Method [2] are also options, but these, like the first method, can encounter problems when faced with states which lie close in mass, or when unphysical contributions appear due to quenching (ghosts).…”
Section: Introductionmentioning
confidence: 99%