data-analysis-physics

Layer 1 — Physics24 concepts in this subtree

Physics data analysis — the application of probability, statistics, and inference to experimental physics data. Distinguished from general machine-learning by two constraints: (i) data-generation mechanism is typically a known (or…

Likelihood ratio: Λ = L(θ_1|x)/L(θ_0|x)
χ²(k) moments: E[χ²]=k, Var[χ²]=2k
Bootstrap distinct-fraction: 1 - (1-1/n)^n → 1 - 1/e
LR null: Λ(L_1=L_0) = 1
χ²(k=1) anchor: mean=1, variance=2
Bootstrap distinct-limit: lim_{n→∞} [1-(1-1/n)^n] = 1-e^{-1}
Least-squares via SVD: beta-hat = V Sigma^+ U^T y; Frobenius-norm extremum
MLE first-order optimality: d ln L/d theta = 0 at theta-hat; Fisher information
AIC for nested models: AIC = 2k - 2 ln L; Hahn-Banach extension on model functionals
Theorem: trace(X X^T) - 13 = 0 at X = diag(2, 3) (SVD Frobenius-norm-squared)
Theorem: d/d mu [-(x - mu)^2/(2 sigma^2)] at mu = x equals 0 (MLE first-order)
Theorem: Delta AIC - 2(k_2 - k_1) = 0 at L_1 = L_2 (equal-likelihood nested-model penalty)
Matched filter (Wiener 1949)
MCMC (Metropolis 1953 / Hastings 1970)
HMC (Duane-Kennedy-Pendleton-Roweth 1987)
Kalman filter (physics applications)
Bayesian model selection (Jeffreys 1939)
AIC / BIC (Akaike 1974, Schwarz 1978)
Least squares (Gauss 1809)
MLE (Fisher 1922)
Bayesian (Laplace 1812)
Kalman filter (1960)
Monte-Carlo (Ulam 1947)
ML for physics (Mehta 2019)
Explore the data-analysis-physics subtree on the interactive graph →