can i get some help on trying to give a 90% confidence interval for the slope of the true regression line.

can i get a step by step "guide"?

3/28/2007 9:17:25 PM

Do you have a table of observations for reference? I remember that you could find confidence intervals for slopes and intercepts using Excel.

3/28/2007 9:28:43 PM

yes i do

3/28/2007 9:31:51 PM

Alright, cool. I suppose if you're out of ideas (or just lazy, like a lot of us), it's worth a shot to use the Data Analysis Toolpak in Excel. It's under "Tools" and "Data Analysis." If you've got it installed, hit "Regression" after the "Data Analysis" window pops up and use that to define your data coordinates in that table and what confidence level you want.

... If you understood any of that, I applaud you. Otherwise, I'll throw you an IM or something and we'll go from there.

3/28/2007 9:36:32 PM

If you are talking about a simple linear regression y = Ax + B, then:

Let N be the count of data points indexed by i, e.g. (Xi,Yi).
Let Sxx = Summation(Xi-aveX)*(Xi-aveX).
Let Sxy = Summation(Xi-aveX)*(Xi-aveX).
Let Se = Summation(Yi-Yest)*(Yi-Yest).

Then A = Sxy/Sxx,
and the two-sided confidence interval for A is A +/- t(0.1/2,N-2)*Se*sqrt(1/Sxx).

3/28/2007 9:37:55 PM

after you complete the task, you should be a smart ass and write in your answer that in business and the social sciences a 95% confidence interval is standard practice (an alpha level of .05 haha).

PS your TI will do this for you

[Edited on March 28, 2007 at 9:42 PM. Reason : .]

3/28/2007 9:39:51 PM

Quote :

"Alright, cool. I suppose if you're out of ideas (or just lazy, like a lot of us), it's worth a shot to use the Data Analysis Toolpak in Excel. It's under "Tools" and "Data Analysis." If you've got it installed, hit "Regression" after the "Data Analysis" window pops up and use that to define your data coordinates in that table and what confidence level you want. ... If you understood any of that, I applaud you. Otherwise, I'll throw you an IM or something and we'll go from there."

i'm goin with this for now, i think i can handle it.

^^my number 2 man.

^yea i have to include my alpha level and interval and all that jazz in my conclusion.

i'm in AP Stats in HS, when did you guys get to this stuff?

[Edited on March 28, 2007 at 9:45 PM. Reason : ]

[Edited on March 28, 2007 at 9:46 PM. Reason : ]

3/28/2007 9:45:17 PM

Quote :

"i'm in AP Stats in HS, when did you guys get to this stuff?"

I took it when I was a sophomore. I knew a few people who took AP Statistics in high school. Most of them seemed to hate it.

3/28/2007 9:50:38 PM

turns out i don't have the data analysis toolpak, if i did i woulduv been done by now

i'll give NatureNut's way a try, it should work.

^yea i took algebraII last year so i went into stats with no prerequisites. i guess it's kuhz of that why i hate it because it doesn't seem THAT hard if i had at least some prior experience with this kind of math.

[Edited on March 28, 2007 at 9:52 PM. Reason : ]

3/28/2007 9:51:06 PM

You prob just have to add it in

Tools--->add ins--->tool pack

If you can install it that way, then you should be able to hit the tools button again and find the data analysis package at the bottom of the tab

3/28/2007 10:59:17 PM

Quote :

"i'm in AP Stats in HS, when did you guys get to this stuff"

highschool stats was about the same as intro stats here at state (at least you don't have to put up with all the dumbfucks who took it at state).

3/28/2007 11:13:54 PM

Quote :

"You prob just have to add it in Tools--->add ins--->tool pack If you can install it that way, then you should be able to hit the tools button again and find the data analysis package at the bottom of the tab "

yep that was it. i got it now.


THANKS FOR THE HELP EVERYONE, especially you guys themetr0/NatureNut

3/28/2007 11:16:42 PM

ST311

talk about boring.

once a week

easy A.

lotsa BS.

3/29/2007 1:25:01 PM

Quote :

"Let Sxy = Summation(Xi-aveX)*(Xi-aveX)."

has a typo.

Since it is summation of x and y, it's Sxy = Summation(Xi-aveX)*(Yi-aveY)

3/29/2007 1:29:17 PM