pub fn bayesian(
observed: Option<&Observations>,
stated: &Stated,
noise: Noise,
forward: &mut dyn Forward,
sampling: Sampling,
) -> BayesianExpand description
The Bayesian option (ADR-317, ADR-320): fit, then the posterior of
its unknowns drawn around its answer by posterior::sample_in_two, each draw a flight flown.
The posterior is the least-squares fit’s own on the heights: the same residuals at the same
noise, the noise widened for a residual that runs on from bin to bin as the fit’s errors widen
it, the same bins read and the same priors. On a track whose noise the fit takes from the log,
it reads the drift through a receiver’s navigation filter and integrates out the track’s noise,
its run and the pad’s position (ADR-321, Problem::track_noise_integrated); its
draws are then centered near that posterior’s peak, by three Gauss-Newton steps. The
sensor’s four linear unknowns (height offset, gain, drift, port) are integrated out in closed
form, −2 ln p(θ) = χ²_min(θ) + ln det H(θ), so seven unknowns are drawn: the six flown and
the clock. On a log of height alone the wind’s shear exponent is an eighth, uniform on
[0, SHEAR_EXPONENT]. The wind is drawn first, from its block of the fit’s covariance at the
answer; the rest about where one Gauss-Newton step puts them in that wind, from their
covariance given the wind; a pilot widens either block’s scale where the posterior reaches
further. Its 90% credible intervals are the posterior’s 5% and 95% points as they come: no
shares widen them.