# Thread: Statistical Analysis and error calculation (Excel 2002)

1. ## Statistical Analysis and error calculation (Excel 2002)

I have recently begun a pysics investigation and have decided to use Excel.I have created my graph of all the data, however for use in a calculation I must know the error percentage in the gradient of the trendline that excell has placed on the graph, does any one know how I can either display this or calculate it.????

2. ## Re: Statistical Analysis and error calculation (Excel 2002)

Check out the Tools, Data analysis option, which gives more control over what you get.

Be aware, that Excels' statistical features are far from precise and sometimes even yield faulty results. It is better to use dedicated statistical software for real life statistical tasks.

3. ## Re: Statistical Analysis and error calculation (Excel 2002)

Yeh I know they are but I really need this thing and I dont have time to get the new software so does n e 1 know a way to get excel to do it or even a formula for working this thing out.

4. ## Re: Statistical Analysis and error calculation (Excel 2002)

As I said:

Tools, Data analysis, regression.

5. ## Re: Statistical Analysis and error calculation (Excel 2002)

Yeh but dosent regression just give you how well the points lie on the line????? Thats not what the percentage error in the gradient is, is it???

6. ## Re: Statistical Analysis and error calculation (Excel 2002)

Hey, why not just try? Look at the attached sheet what it can do!

I am pretty sure what you need is there.

7. ## Re: Statistical Analysis and error calculation (Excel 2002)

k thanks m8t I know ive bin a pain but uve saved my @#\$!% big time.

8. ## Re: Statistical Analysis and error calculation (Excel 2002)

If you apply the regression tool from the Analysis Toolpak, then the standard error of the regression line is given. This standard error of the regression line is simply the root of (the sum of squares of residuals divided by the number of degrees of freedom df). df = n-2 in case of linear regression and n is the number of observations.

By the way, regression in Excel can be trusted, except for the adjusted R-square when fitting through the origin (zero intercept); then the calculation of the adjusted R-square is wrong. For the rest, you can trust the results.

I don't know what you exactly mean with the error percentage in the gradient of the trendline, but there exists a way to calculate confidence intervals and even prediction intervals for a regression line. Is that what you need or did Jan Karel solve your problem?

9. ## Re: Statistical Analysis and error calculation (Excel 2002)

yeh dude i think thats it if u can tell me PLEASE ill check when i get back to collage [img]/forums/images/smilies/smile.gif[/img]

10. ## Re: Statistical Analysis and error calculation (Excel 2002)

Now for the LAST TIME!

Check out the sheet I included some 100 posts ago.

Look at the range a26:G28. It shows all statistics needed, including the confidence interval of the intercept.

Sorry for loosing my patience.

11. ## Re: Statistical Analysis and error calculation (Excel 2002)

Tokra, you will receive well thought out and well written replies on this board. In turn we would ask that you spend some time phrasing your questions in proper English, looking at the replies, and doing some work yourself. You have just irritated two geniuses at both statistics and Excel, provoked those of us whose first language is English, and confused those who are struggling with English. I would like to learn more about statistical analysis, not just be annoyed by your lack of courtesy. What are you learning at

12. ## Re: Statistical Analysis and error calculation (Excel 2002)

<<geniuses>>

You didn't mean me, did you <vbg>?

13. ## Re: Statistical Analysis and error calculation (Excel 2002)

Yes I did and I'll stick by it. Microsoft doesn't give out MVP stickers as Halloween treats. I

14. ## Re: Statistical Analysis and error calculation (Excel 2002)

I never took a MOUS exam. Wouldn't know for the world if I'd pass <g>.

15. ## Re: Statistical Analysis and error calculation (Excel 2002)

The core is a snap, but the expert is a bear: everything from scenarios to database queries.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•