Table 2 Best subset model regression coefficients and p values of selected variables using a significance level of 0.1.

(Intercept)

TRUE

1.05839E−01^***

P_A

TRUE

2.69350E−02^***

λ

TRUE

−2.08196E−03^***

μ

TRUE

−9.61593E−04^***

γ

TRUE

1.14047E−04^*

−dγ/dT

TRUE

2.26852E−04^***

\({P}_{\rm{A}}\,*\,\lambda\)

FALSE

0

\({P}_{A}\,*\,\mu\)

TRUE

4.53708E−04^***

\({P}_{\rm{A}}\,*\,\gamma\)

TRUE

2.44496E−04^***

\({P}_{\rm{A}}\,*\, - \, {\rm{d}}\gamma/{\rm{dT}}\)

TRUE

3.08903E−04^***

\(\lambda\,*\,\mu\)

TRUE

−3.96380E−04^***

\(\lambda \, *\, \gamma\)

TRUE

−3.77203E−04^***

\(\lambda\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

−5.63809E−04^***

\(\mu\,*\,\gamma\)

FALSE

0

\(\mu\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

FALSE

0

\(\gamma\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

FALSE

0

\({P}_{\rm{A}}\,*\,\lambda\,*\,\mu\)

TRUE

1.86421E−04^**

\({P}_{\rm{A}}\,*\,\lambda\,*\,\gamma\)

FALSE

0

\({P}_{\rm{A}}\,*\,\lambda\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

6.75741E−04^***

\({P}_{\rm{A}}\,*\,\mu\,*\,\gamma\)

TRUE

−1.55266E−04^**

\({P}_{\rm{A}}\,*\,\mu\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

2.25628E−04^***

\({P}_{\rm{A}}\,*\,\gamma\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

2.24396E−04^***

\(\lambda\,*\,\mu\,*\,\gamma\)

TRUE

2.93934E−04 ^***

\(\lambda\,*\,\mu\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

2.92017E−04^***

\(\lambda\,*\,\gamma\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

−1.42671E−04^**

\(\mu\,*\,\gamma\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

−3.14491E−04^***

\({P}_{\rm{A}}\,*\,\lambda\,*\,\mu\,*\,\gamma\)

TRUE

2.22234E−04^***

\({P}_{\rm{A}}\,*\,\lambda\,*\,\mu\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

−2.71283E−04^***

\({P}_{\rm{A}}\,*\,\lambda\,*\,\gamma\,*\,-{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

8.13880E−04^***

\({P}_{\rm{A}}\,*\,\mu\,*\gamma\,*\,-\,{\rm{d}}\gamma/{\rm{dT}}\)

TRUE

−1.59792E−04^**

\(\lambda\,*\,\mu\,*\,\gamma\,*\,-\,{\rm{d}}\gamma/{\rm{dT}},\)

TRUE

1.05624E−03^***

Adj. R²

0.99988206

^***p < 0.01; ^**p < 0.05; ^*p < 0.1.

Standard error = 0.00005287

for all coefficient estimates