PDA

View Full Version : Question regarding F and T Statistic, std error?


dr.asma
01-24-2009, 06:57 PM
Question regarding F and T Statistic, std error?
Kindly let me know about this equation, how it is solved?

this is the equation:
The quadratic regression model is fitted to the data. The quadratic
model is given as:
GDP =  + β1 ODA + β2 ODA2
The results of the estimated model are given as:
GDP = -4371.2492 + 64.4 ODA - 0.014 ODA2
Coefficient Data:
Overall Significance of model (F – Statistic) = 24.92458606177 *
 = -4371.2492
β1 = 64.402631 (t statistic = 3.48)*
β2 = -0.013780676 (t statistic = -1.69)**
R squared = 0.579068 = 58%
Standard Error = 14794.9028265
Correlation Coefficient (r) = 0.7609605 = 76%

* Significant at 1% level of significance
** Significant at 10% level of significance

The box in the equation is actaully alpha
The ODA2 means ODA raised to the power 2

and the data on which it is fitted is as follows:

http://mpra.ub.uni-muenchen.de/1211/1/MP...

please look at the table 4 on page 16

Ummmm quick replies are appreciated!

thanks

quark
01-24-2009, 10:30 PM
From the t-statistics, β1 is significant at 1% level of significance, β2 is significant at 10% level of significance.


From the F-statistic, the model is significant at 1% level of significance.

SE = 14795

dr.asma
01-25-2009, 02:27 AM
can you please explain how you met this conclusion, thanks

quark
02-06-2009, 11:20 PM
The results are listed in the output.

(t statistic = 3.48)* the one star indicates that β1 is significant at 1% level of significance.

(t statistic = -1.69)** the two stars indicate that β2 is significant at 10% level of significance.

Standard Error = 14794.9028265 = 14795 approx.

dr.asma
03-09-2009, 02:38 AM
that i do understand my friend! thanks

But my problem really is that: how can you verify the value of

alpha

beta 1

beta 2

please explain the quadratic model in detail!


thankyou

any help is much appreciated!