Advertisement
Help Keep Boards Alive. Support us by going ad free today. See here: https://subscriptions.boards.ie/.
If we do not hit our goal we will be forced to close the site.

Current status: https://keepboardsalive.com/

Annual subs are best for most impact. If you are still undecided on going Ad Free - you can also donate using the Paypal Donate option. All contribution helps. Thank you.
https://www.boards.ie/group/1878-subscribers-forum

Private Group for paid up members of Boards.ie. Join the club.

Econometrics exam question (fairly urgent!)

  • 07-12-2010 09:32PM
    #1
    Registered Users, Registered Users 2 Posts: 66 ✭✭


    Hey I've an econometrics exam tomorrow. There's a question thats says.. you are given following data on 15 observations and gives Ex2i squared value, Exiyi value etc..

    I can estimate the intercept and slope coefficient and work through it until you get the t value (beta over S.E) but in the solution (question is for 95% C.I) its says..

    T= 3.4974 therefore Beta 1 is significant (same situation with Beta 2 with 14 or something)

    Anyway how do you decide if its significant, is it if its not between -1.96 and + 1.96 because of the 95% C.I?


    Another similar question, in ramseys RESET I found the F m;n,k value which was 4, and I know that if this is greater than the critical F value the model is misspecified, but how do I find the F value on the tables such as this (95% again)..

    http://home.comcast.net/~sharov/PopEcol/tables/f005.html

    Theres six parameters in the unrestricted and 3 in the restriced model so is it 3,6 or something?

    Sorry if that was a bit long winded!

    Thanks


Comments

  • Closed Accounts Posts: 2,208 ✭✭✭Économiste Monétaire


    The t-statistic is significant at the 1% level. The t = 3.4974 is from testing if [latex] \displaystyle \beta_{1} = 0 [/latex], i.e.,

    [latex] \displaystyle \frac{\beta_{1} - 0}{se(\beta_{1})} = 3.4974 [/latex]

    Look this t-stat up in your tables, with df = 12.

    For the second question, can you give the values for k, m, and n?


  • Registered Users, Registered Users 2 Posts: 66 ✭✭lopppy


    Little confused over using the tables, do i go to 12 on the df and .1 then on the top line, getting 1.782. So is it because the 4.5 t value is greater than this 1.782 that its significant?


    M N K are 3, 44, 6 respectively. Dont need the answer to that question specifically, just how do look up the F value?


  • Closed Accounts Posts: 2,208 ✭✭✭Économiste Monétaire


    .1 means 10% significance level. If that's your cut-off, then, yes, it is significant. Move over to the 0.05 part, and you'll see 2.179, which is less than 3.4974, so it's also significant at the 5% level.

    For the F value, the row of red numbers at the top of your link is the first part in [latex]\displaystyle F_{m,(n-k)} [/latex], which is 3, and the vertical list of red numbers is for second part, n-k, which is 38. The critical value is therefore between 2.84 and 2.87, at the 5% significance level.


  • Registered Users, Registered Users 2 Posts: 66 ✭✭lopppy


    .1 means 10% significance level.

    You mean 90%? :p



    Thanks though you saved me staying up half the night!


  • Closed Accounts Posts: 2,208 ✭✭✭Économiste Monétaire


    Most people just say 10%/5%/1% significance level referring to P>|t|. Best of luck with the test.


  • Advertisement
Advertisement