Table 3 Electron density models and statistical measures.

From: Probing the solar corona with very long baseline interferometry

Session

Reduced χ 2

ΔAIC

Relative likelihood

N0 (1012 m−3)

RD1106

1.1

−5.1

0.08

0.0±0.4

RD1107

1.1

−49.5

0.00

1.7±0.4

RD1201

1.7

−8.4

0.02

3.3±1.3

RD1202

1.2

−0.3

0.85

0.9±0.4

RD1203

1.2

−2.2

0.33

2.2±1.8

RD1204

0.8

−6.4

0.04

1.2±1.0

RD1205

0.8

−4.1

0.13

0.5±0.3

RD1206

0.5

−8.4

0.02

0.3±0.1

RD1207

2.4

0.9

0.65

0.6±0.8

RD1208

2.0

−64.2

0.00

1.5±0.4

RD1209

0.6

−2.8

0.25

0.1±0.3

RD1210

0.6

−16.8

0.00

2.5±0.6

Weighted mean over all R&D sessions

0.57±0.18

  1. For each of the IVS R&D sessions in 2011 and 2012, a least-squares adjustment was performed estimating the parameters of the Sun’s corona, the Earth’s ionosphere and instrumental delays. The goodness of fit is indicated by the reduced χ2. Another adjustment without estimating the electron density parameter N0 was performed and for both approaches, the Akaike Information Criterion (AIC) was computed. The difference ΔAIC=AICcorona–AICwithout is given as well as the relative likelihood exp(ΔAIC/2) of the model excluding the solar corona with respect to the one which takes the corona into account (vice versa for RD1207). Furthermore, the estimated parameters N0 of the electron density model and their 1σ s.e. from the least-squares adjustment (obtained from the a posteriori variance-covariance matrix) assuming β=2 are listed.
Back to article page