#8636 A regression equation with 4 independent variables is estimated using 20 data points. The R2 is 0.46. An analyst is testing to see whether all of the coefficients are equal to zero. The p-value for the test is: A) lower than 0.025. B) between 0.05 and 0.10. C) greater than 0.10. D) between 0.025 and 0.05.
steps: assume total variance = 1, explained variance = 0.46, unexplained = 1-0.46 = 0.54 mean explained = 0.46/4 = 0.115 mean unexplained = 0.54/(20-4-1)= .036 F statistic = 3.195 -> look up in the table F(15,4) that value of 3.195
I don’t get it… :)…so whats the answer again?
Schweser says the answer is not C… try again
I think the F-Statistic is 3.06 but I don’t know how to get the P-value.
Gotta use F-Tables…
ok, i need more studying… although i prefer these types of questions (tying things together) than vague wording on the question.
monki Wrote: ------------------------------------------------------- > Schweser says the answer is not C… try again I don’t care what Schweser says. if I messed up somewhere in calculations, I’d like to you to point out where.
maratikus looks right to me right up until the end. We know that explained is .46 so unexplained has to be .54 so the total is equal to 1.00 (100%). So now you know RSS=.46 and SSE=.54 and can use that to calculate F: MSR/MSE = (.46/4)/(.54/(20-(4+1))=3.19444 P is the greatest significance value at which the null can be rejected if I recall correctly. My F table at 2.5 percent say at 4,15 that the critical value is 3.80 and at 5% it is 3.06 (remember out value is 3.19444. So we now know that the null could be rejected at 5% but not at 2.5%. Answer should be D.
It’s the critical value of F that’s wrong. It should be F(4,15), not F(15,4) Numerator is regression, which has 4 df, denominator is error, which has 15 df. So using http://davidmlane.com/hyperstat/F_table.html you get p = 0.04397
excellent! thanks mwvt9 and chrismaths.
The answer is D… i didn’t point out why you got it wrong bcoz I am trying to figure out too lol… looks like somebody got it right… good job!!
Ok. Right answer in full: [maratikus] steps: assume total variance = 1, explained variance = 0.46, unexplained = 1-0.46 = 0.54 mean explained = 0.46/4 = 0.115 mean unexplained = 0.54/(20-4-1)= .036 F statistic = 3.195 -> look up in the table [/maratikus][chrismaths] F(4,15) [/chrismaths][mwvt9] My F table at 2.5 percent say at 4,15 that the critical value is 3.80 and at 5% it is 3.06 Answer is D [/mwvt9] Alles klaar?