# How do we compute DF and T-critical (Exampe #2)

Wanda Brunner, CFA, is trying to calculate a 99% confidence interval for a regression equation based on the following information: What are the lower and upper bounds for variable CS? The coefficient for CS is .32 The standard error is .025 My first question is: How do we know it’s a two tailed test? My second question is: How do we know what the DF are so we can compute T-critical? This is what they give as the answer: The critical t-value is 2.42 at the 99% confidence level (two tailed test). The estimated slope coefficient is 0.32 and the standard error is 0.025. The 99% confidence interval is 0.32 ± (2.42)(0.025) = 0.32 ± (0.061) = 0.260 to 0.381. ~~~ From the first example with a 95% confidence level, T-crit was 2.02, which has about 40 df. From this example which is a 99% level, the T-crit is 2.42, which has more that 120 df. How do we compute DF???

1. two tailed - it is because they ask for upper and lower bounds. 2. your usual culprit for degrees of freedom would be n - (k + 1) where k = number of independent variables. and n=# of data points used for the regression analysis.

CPK Can you please compute DF from the data given above? thanks

I cannot compute it from the data above. you have put in a partial print of data. computing the data and going backwards is absolutely nonsense. it will be provided in the ANOVA table or the source of the question. post the full question… and I can then try.

First, thanks for responding! Second, how do we compute DF given the data below? Thanks Wanda Brunner, CFA, is trying to calculate a 99% confidence interval for a regression equation based on the following information: …coefficient…standard error intercept…-10.6%…1.357 DR… .52 … .023 CS… .32 … .025 What are the lower and upper bounds for variable CS? A) 0.260 to 0.381. B) 0.267 to 0.374. C) 0.274 to 0.367. note: qbank # 117082

they should’ve given you the number of sample points… haven’t they? in the absence of that - you cannot do this …

Thanks. I understand that DF is basically sample size - n-(k+1) but they never gave us n is these two examples, and I was wondering why I was totally lost. I didn’t leave anything out of the data that I posted. Maybe that information is on a different question.

man - it gives you df=40 on the question… at least when I look at the question online

also the online version says 98% confidence interval (df=40). possibly an update of QBank is called for.

I use the qbank, but not the online qbank. I use the qbank that is downloaded to the hard-drive. On my question, the df is not given