The t statistic and the partial F statistic are closely related. ...

[schs14]

The t statistic and the partial F statistic are closely related. Explain this relationship. Which would you prefer to use?
 
  What will be an ideal response?

Question 2

The probability of rejecting a null hypothesis when the null hypothesis is true is known as the:
 
  A) confidence level.
  B) p-value.
  C) power of the test.
  D) significance level.

[kristenb95]

Answer to Question 1

In multiple regression analysis, the t statistic is used to test if an individual population slope coefficient is significantly different from zero  where the coefficient is the change in the dependent variable given a unit change in the independent variable, while all other explanatory variables are held constant. Essentially, it tells us whether that particular explanatory variable makes a significant contribution to the model. The partial F statistic is used to test if an individual explanatory variable or a set of individual explanatory variables is significant to the model, holding all other variables fixed. So, for the individual problem, the F statistic and the t statistic must produce identical results and they do, because t2 = F in these cases. As you might guess, t is more widely used because it is much easier to calculate and access from statistical packages.

Answer to Question 2

D

[schs14]

Thanks for the timely response, appreciate it

[nguyenhoanhat]

Gracias!