Table 3
Coefficients of the reduced regressive models.
| β | Additional information | RR1 | RR2 | RR3 | RR4 | RR5 |
|
|
||||||
| β 0 | intercept | 39.211 | 46.545 | 45.791 | 38.112 | 43.060 |
| β 1 | T eff | −214.985 | −215.697 | −219.870 | −200.457 | −197.401 |
| β 2 | log g | 4.350 | −0.195 | 0.518 | 0.591 | −2.013 |
| β 3 | [Fe/H] | −0.444 | 0.964 | −3.285 | −6.774 | −1.320 |
| β 4 | Tefflog g | −12.713 | 7.492 | 6.000 | 3.889 | 8.912 |
| β 5 | Teff [Fe/H] | 5.651 | −2.778 | 5.654 | 9.476 | 3.672 |
| β 6 | log g [Fe/H] | −2.035 | −1.044 | −1.049 | −0.715 | −1.287 |
| β 7 | Teff2 | 325.152 | 274.256 | 285.581 | 272.625 | 253.623 |
| β 8 | log 2g | −0.003 | −0.972 | −0.979 | −0.705 | −0.692 |
| β 9 | [Fe/H] 2 | −0.832 | −1.151 | −1.688 | −2.791 | −1.022 |
| β 10 | Teff2log g | 7.427 | −8.709 | −8.170 | −6.970 | −9.380 |
| β 11 | Teff2 [Fe/H] | −3.282 | 4.068 | 0.963 | 0.000 | 1.359 |
| β 12 | Tefflog 2g | 0.170 | 1.050 | 1.082 | 1.025 | 1.005 |
| β 13 | log g2 [Fe/H] | −0.057 | −0.040 | −0.033 | −0.025 | −0.030 |
| β 14 | Teff [Fe/H] 2 | 1.805 | 1.804 | 3.009 | 3.339 | 2.656 |
| β 15 | log g [Fe/H] 2 | 0.000 | 0.000 | −0.062 | 0.000 | −0.102 |
| β 16 | Tefflog g [Fe/H] | 2.698 | 1.470 | 1.297 | 0.976 | 1.455 |
| β 17 | T eff 3 | −101.789 | −56.074 | −61.606 | −59.915 | −48.618 |
| β 18 | log 3g | 0.000 | 0.019 | 0.018 | 0.000 | 0.000 |
| β 19 | [Fe/H] 3 | 0.116 | 0.000 | 0.143 | −0.117 | 0.220 |
|
|
||||||
| σ FIT | × 105 erg cm-2 s-1 Å-1 | 0.048(1.2%) | 0.059(1.2%) | 0.065(1.5%) | 0.073(1.8%) | 0.068(1.7%) |
| R 2 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | |
| ΔBIC | BIC(M) − BIC(MR) | 11.2 | 11.1 | 0.0 | 15.9 | 4.7 |
| Removed coefficients | β18, β15 | β19,β15 | none | β18, β15, β11 | β 18 | |
Notes. First and second columns are the coefficients of the model describing the reference regions and additional information regarding the first column. The remaining columns show, for each RR, the values of the corresponding coefficient described in first and second columns (following Eq. (3)). Respectively, the last four rows are: fitting errors in 105 erg cm-2 s-1 Å-1 units, R2 multiple correlation coefficient, BIC difference between complete, and final reduced cubic regressive models, and the variables that were removed during the Stepwise regression. The percentages that are shown together with the fitting errors are its fraction compared to typical solar chromospheric radiative losses derived in Sect. 5. RR3 did not suffer any removal, reason why the BIC difference is zero.