Error of slope taking into account error of the data points
Hi!
I need to find the error of the slope of the line of best fit taking into account the error of the individual data points. I have a fixed value for the error of the points that I have figured out independently. (It's about .8 for every point) I have the error bars on, but excel isn't taking the error of the points into account when determining the error of the slope! So...how do I solve this? I don't necessarily need an excel function to do it for me, a formula would be fine. Thanks! |
Error of slope taking into account error of the data points
The slope of a line is given by the change in y values divided by the
change in corresponding x values, i.e.: slope = (y2 - y1) / (x2 - x1) where x2,y2 is at one end of the line and x1,y1 is at the other end. In your case the errors in the y values are plus or minus 0.8, so in the worst two cases y2 may be +0.8 with y1 being -0.8, or y2 may be -0.8 and y1 may be +0.8, so you could work out the two extremes from this. Hope this helps. Pete On Jul 7, 5:17*pm, cer144 wrote: Hi! I need to find the error of the slope of the line of best fit taking into account the error of the individual data points. I have a fixed value for the error of the points that I have figured out independently. (It's about .8 for every point) I have the error bars on, but excel isn't taking the error of the points into account when determining the error of the slope! So...how do I solve this? I don't necessarily need an excel function to do it for me, a formula would be fine. Thanks! |
Error of slope taking into account error of the data points
Thanks!! I'll try that, see if it's good enough for the boss lol.
"Pete_UK" wrote: The slope of a line is given by the change in y values divided by the change in corresponding x values, i.e.: slope = (y2 - y1) / (x2 - x1) where x2,y2 is at one end of the line and x1,y1 is at the other end. In your case the errors in the y values are plus or minus 0.8, so in the worst two cases y2 may be +0.8 with y1 being -0.8, or y2 may be -0.8 and y1 may be +0.8, so you could work out the two extremes from this. Hope this helps. Pete On Jul 7, 5:17 pm, cer144 wrote: Hi! I need to find the error of the slope of the line of best fit taking into account the error of the individual data points. I have a fixed value for the error of the points that I have figured out independently. (It's about .8 for every point) I have the error bars on, but excel isn't taking the error of the points into account when determining the error of the slope! So...how do I solve this? I don't necessarily need an excel function to do it for me, a formula would be fine. Thanks! |
didn't work!
Didn't work lol! It just changed the y-intercept!! Still need some help on
this one. |
Error of slope taking into account error of the data points
Another thing about errors is that if you combine two (or more) values
with addition or subtraction to get a result, then the error in the result is the sum of the errors in the two (or more) components. However, if you combine using multiplication or division, then the %age errors are added to give the cumulative error in the result. Hope ths helps. Pete On Jul 7, 6:58*pm, cer144 wrote: Thanks!! I'll try that, see if it's good enough for the boss lol. "Pete_UK" wrote: The slope of a line is given by the change in y values divided by the change in corresponding x values, i.e.: slope = (y2 - y1) / (x2 - x1) where x2,y2 is at one end of the line and x1,y1 is at the other end. In your case the errors in the y values are plus or minus 0.8, so in the worst two cases y2 may be +0.8 with y1 being -0.8, or y2 may be -0.8 and y1 may be +0.8, so you could work out the two extremes from this. Hope this helps. Pete On Jul 7, 5:17 pm, cer144 wrote: Hi! I need to find the error of the slope of the line of best fit taking into account the error of the individual data points. I have a fixed value for the error of the points that I have figured out independently. (It's about .8 for every point) I have the error bars on, but excel isn't taking the error of the points into account when determining the error of the slope! So...how do I solve this? I don't necessarily need an excel function to do it for me, a formula would be fine. Thanks!- Hide quoted text - - Show quoted text - |
didn't work!
What do you mean "it didn't work"? - I didn't tell you to do anything
in Excel. You have a (value for the) slope, so choose two points on this line to give you the x1,y1 and x2,y2 values. Then using the (mathematical) formula that I gave, you can work out a value for the slope for one extreme, and then a value for the slope for the other extreme - these represent the range in values that the slope can take, and you can express that as an error range for the value of the slope that you start with. Hope this helps. Pete On Jul 7, 7:10*pm, cer144 wrote: Didn't work lol! It just changed the y-intercept!! Still need some help on this one. |
All times are GMT +1. The time now is 08:39 PM. |
Powered by vBulletin® Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
ExcelBanter.com