Skip to content

Instantly share code, notes, and snippets.

@MarinMario
Last active January 22, 2026 13:45
Show Gist options
  • Select an option

  • Save MarinMario/565cce2366a0b304513c6f373ff0e267 to your computer and use it in GitHub Desktop.

Select an option

Save MarinMario/565cce2366a0b304513c6f373ff0e267 to your computer and use it in GitHub Desktop.
t-stat = coefficient/std-error
|t-stat| > t-crit => prob < 0.05 => statistic semnificativ
F-stat = (R^2/k)/(1-R^2)(n-k-1) (k = nr variabile explicative, n = nr observatii)
F-stat > F-crit => model valid global pt un nivel de semnificatie de 5%
R^2 = coeficient de determinare, ex: R^2 = 0.7 => modelul explica 70% din rezultat, restul de 30% sunt variabile pe care modelul nu le ia in calcul
Interval incredere = valoarea coeficientului +- tcrit * eroare standard
Tabelul Variance Inflation Factors (Eviews)
variance inflation factors = tabel de verificare coliniaritate
vif < 5 => nu e coliniaritate
vif > 5 => coliniaritate moderata => coeficientul poate fi afectat
vif > 10 => coliniaritate servera => coeficientul e instabil, variabila trebuie eliminata
Tabelul Heteroskedasticity Test: White (Eviews)
homoscedascitate = variatia erorilor este constanta
heteroscedascitate = variatia erorilor nu este constanta, deci rezultatele t-stat pot fi afectate.
prob < 0.05 = heteroscedascitate
prob > 0.05 = homoscedascitate
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment