ECON
LINREG2

Although a t value is a test statistics of the null

• Notes
• 29

This preview shows pages 12–16. Sign up to view the full content.

Although a t-value is a test statistics of the null hypothesis that the corresponding coefficient in the regression model is zero, it is quite easy to rebuild the t-value for testing other null hypotheses, as follows. Suppose you want to test the null hypothesis that where β ' β 0 , β 0 is a given number, for example Then β 0 ' 1.

This preview has intentionally blurred sections. Sign up to view the full version.

13 \$ \$ & \$ 0 \$ F \$ \$ ' \$ \$ \$ F \$ \$ & \$ 0 \$ F \$ \$ ' \$ \$ \$ F \$ \$ & \$ 0 \$ \$ \$ \$ \$ F \$ \$ ' \$ \$ \$ F \$ \$ 1 & \$ 0 \$ \$ ' \$ \$ & \$ 0 \$ \$ . \$ t \$ \$ , (21) so that by Proposition 5, \$ t \$ \$ , \$ ' \$ 0 ' \$ \$ & \$ 0 \$ \$ . \$ t \$ \$ - t n & 2 . (22) For example, suppose that in the ice cream case we want to test the null hypothesis . H 0 : β ' 1 Then \$ t \$ \$ , \$ ' 1 ' \$ \$ & 1 \$ \$ . \$ t \$ \$ ' 1.5 & 1 1.5 ×4.597 . 1.532, (23) which under the null hypothesis is a random drawing from the t distribution with 6 H 0 : β ' 1 degrees of freedom. Note that the value of this test statistic is in the acceptance regions in Figures 2 and 3. This trick is useful if the econometric software you are using only reports the t-values but not the standard errors. If the standard errors are reported, you can compute directly as ˆ t ˆ β , β ' β 0 Of course, if only the standard errors are reported and not the t-values you ˆ t ˆ β , β ' β 0 ' ( ˆ β & β 0 )/ˆ σ ˆ β . can compute the t-value of as ˆ β ˆ t ˆ β ' ˆ β / ˆ σ ˆ β . 6. The R 2 The R 2 of a regression model compares the sum of squared residuals ( SSR ) of the model with the SSR of a “regression model” without regressors: Y j ' " % U j , j ' 1,2, ..... , n . (24) It is easy to verify that the OLS estimator of α is just the sample mean of the ‘s: ˜ α Y j # " ' ¯ Y ' 1 n j n j ' 1 Y j .
14 Therefore, the SSR of “regression model” (24) is which is called the Total Sum of ' n j ' 1 ( Y j & ¯ Y ) 2 , Squares ( TSS ), is TSS ' j n j ' 1 ( Y j & ¯ Y ) 2 . (26) The R 2 is now defined as: R 2 ' def . 1 & SSR TSS . (27) The R 2 is always between zero and one, because SSR # TSS . ( Exercise : Why?) If SSR = TSS , so that R 2 = 0, then model (24) explains the dependent variable ‘s equally well as model (2). In Y j other words, the explanatory variables in (2) do not matter: β = 0. The other extreme case is X j where R 2 = 1, which corresponds to SSR = 0. Then the dependent variable in model (2) is Y j completely explained by , without error: / Thus, the R 2 measures how well the X j Y j α % β X j . explanatory variables are able to explain the corresponding dependent variables For X j Y j . example, in the ice cream case, SSR = 11.5 and TSS = 52, hence R 2 = 0.778846. Loosely speaking, this means that about 78% of the variation of ice cream sales can be explained by the variation in temperature. 7. Presenting regression results When you need to report regression results you should include, next to the OLS estimates of course, either the corresponding t-values or the standard errors, the sample size n , the standard error of the residuals (SER), and the R 2 , because this information will enable the reader to judge your results. For example, our ice cream estimation results should be displayed as either Sales ' & 0.25 % 1.5 Temp ., n ' 8, SER ' 1.384437, R 2 ' 0.778846 ( & 0.100) (4.597) ( t & values between brackets ) or

This preview has intentionally blurred sections. Sign up to view the full version.

15 Sales ' & 0.25 % 1.5 Temp ., n ' 8, SER ' 1.384437, R 2 ' 0.778846 (2.49583) (0.32632) ( standard errors between brackets ) It is helpful to the reader if you would indicate whether you have displayed the t-values between brackets or the standard errors, but you only need to mention this once.
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern