I am doing a regression on real data to find the right formula. The data is as follows:
Wall Real time
2.2 3
2.4 3.3
2.6 3.6
2.8 3.9
3 4.2
3.2 4.5
3.4 4.9
3.6 5.35
3.8 5.8
4 6.35
4.2 6.8
Excel comes up with a formula: y= 0.0153x2+0.1931x+2.8388, The R2 is 0.9991. When I compute my data with the formula these are the resulting figures:
Calc. time
3.37494
3.43472
3.49634
3.5598
3.6251
3.69224
3.76122
3.83204
3.9047
3.9792
4.05554
Not at all like the real data. I tried other trendlines but none works. What am I doing wrong? I did this with some other data and it worked fine.

Thanks for your help.

Ottmar

Hi Ottmar,

Am Wed, 23 Sep 2015 11:21:33 -0700 (PDT) schrieb BottleMan:

I am doing a regression on real data to find the right formula. The data is as follows:
Wall Real time
2.2 3
2.4 3.3
2.6 3.6
2.8 3.9
3 4.2
3.2 4.5
3.4 4.9
3.6 5.35
3.8 5.8
4 6.35
4.2 6.8
Excel comes up with a formula: y= 0.0153x2+0.1931x+2.8388, The R2 is 0.9991. When I compute my data with the formula these are the resulting figures:
Calc. time
3.37494
3.43472
3.49634
3.5598
3.6251
3.69224
3.76122
3.83204
3.9047
3.9792
4.05554
Not at all like the real data. I tried other trendlines but none works. What am I doing wrong? I did this with some other data and it worked fine.

you get the exactest trendline with a XY chart. I got the formula
y=1,2416*e^0,4058*X
and with this formula the Calc time looks like this:
3,0318
3,2881
3,5661
3,8676
4,1946
4,5492
4,9339
5,3510
5,8034
6,2940
6,8262



Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

Hi Ottmar,

Am Wed, 23 Sep 2015 20:38:48 +0200 schrieb Claus Busch:

you get the exactest trendline with a XY chart. I got the formula
y=1,2416*e^0,4058*X
and with this formula the Calc time looks like this:
3,0318
3,2881
3,5661
3,8676
4,1946
4,5492
4,9339
5,3510
5,8034
6,2940
6,8262

please look he
https://onedrive.live.com/redir?resi...=folder%2cxlsm
for "Trendline"


Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

On Wednesday, September 23, 2015 at 2:21:38 PM UTC-4, BottleMan wrote:
I am doing a regression on real data to find the right formula. The data is as follows:
Wall Real time
2.2 3
2.4 3.3
2.6 3.6
2.8 3.9
3 4.2
3.2 4.5
3.4 4.9
3.6 5.35
3.8 5.8
4 6.35
4.2 6.8
Excel comes up with a formula: y= 0.0153x2+0.1931x+2.8388, The R2 is 0.9991. When I compute my data with the formula these are the resulting figures:
Calc. time
3.37494
3.43472
3.49634
3.5598
3.6251
3.69224
3.76122
3.83204
3.9047
3.9792
4.05554
Not at all like the real data. I tried other trendlines but none works. What am I doing wrong? I did this with some other data and it worked fine.

Thanks for your help.

Ottmar

Thanks Claus, this worked. How did you get the formula? When I go to exponential trendline Excel comes up with: 2.7952e^0.0812x. Quite different from yours and only works for the lower numbers.

Ottmar

Hi Ottmar,

Am Thu, 24 Sep 2015 06:20:41 -0700 (PDT) schrieb BottleMan:


How did you get the formula? When I go to exponential trendline Excel comes up with: 2.7952e^0.0812x. Quite different from yours and only works for the lower numbers.

just like you. I went to exponential trendline. Perhaps you have more
data or other data?


Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

On Wednesday, September 23, 2015 at 2:21:38 PM UTC-4, BottleMan wrote:
I am doing a regression on real data to find the right formula. The data is as follows:
Wall Real time
2.2 3
2.4 3.3
2.6 3.6
2.8 3.9
3 4.2
3.2 4.5
3.4 4.9
3.6 5.35
3.8 5.8
4 6.35
4.2 6.8
Excel comes up with a formula: y= 0.0153x2+0.1931x+2.8388, The R2 is 0.9991. When I compute my data with the formula these are the resulting figures:
Calc. time
3.37494
3.43472
3.49634
3.5598
3.6251
3.69224
3.76122
3.83204
3.9047
3.9792
4.05554
Not at all like the real data. I tried other trendlines but none works. What am I doing wrong? I did this with some other data and it worked fine.

Thanks for your help.

Ottmar

No, I used the exact data as shown. And I have done this before without problems. Not sure what is going on. Thank you much for your help.

Ottmar

On Wednesday, September 23, 2015 at 2:21:38 PM UTC-4, BottleMan wrote:
I am doing a regression on real data to find the right formula. The data is as follows:
Wall Real time
2.2 3
2.4 3.3
2.6 3.6
2.8 3.9
3 4.2
3.2 4.5
3.4 4.9
3.6 5.35
3.8 5.8
4 6.35
4.2 6.8
Excel comes up with a formula: y= 0.0153x2+0.1931x+2.8388, The R2 is 0.9991. When I compute my data with the formula these are the resulting figures:
Calc. time
3.37494
3.43472
3.49634
3.5598
3.6251
3.69224
3.76122
3.83204
3.9047
3.9792
4.05554
Not at all like the real data. I tried other trendlines but none works. What am I doing wrong? I did this with some other data and it worked fine.

Thanks for your help.

Ottmar

I use Excel 2007. Maybe that is the problem. I just repeated the entire operation and the result is the same.

Ottmar

Hi Ottmar,

Am Thu, 24 Sep 2015 11:14:00 -0700 (PDT) schrieb BottleMan:

I use Excel 2007. Maybe that is the problem. I just repeated the entire operation and the result is the same.

no, I use it too. The workbook I provided is created in xl2007


Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

Hi Ottmar,

Am Thu, 24 Sep 2015 11:10:29 -0700 (PDT) schrieb BottleMan:

No, I used the exact data as shown. And I have done this before without problems. Not sure what is going on. Thank you much for your help.

are your values calculated or have more decimals as shown in your
question?


Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

Hi Ottmar,

Am Thu, 24 Sep 2015 11:14:00 -0700 (PDT) schrieb BottleMan:

I use Excel 2007. Maybe that is the problem. I just repeated the entire operation and the result is the same.

the best and exactest trendlines you get with XY (scatter) charts.
I made some test and I guess you have a line chart.


Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

Hi again,

Am Thu, 24 Sep 2015 20:42:08 +0200 schrieb Claus Busch:

the best and exactest trendlines you get with XY (scatter) charts.
I made some test and I guess you have a line chart.

from Office help:

If you add a trendline to a line, column, area, or bar chart, the
trendline is calculated based on the assumption that the x values are 1,
2, 3, 4, 5, 6, etc.. This assumption is made whether the x-values are
numeric or text. To base a trendline on numeric x values, you should use
an xy (scatter) chart.


Regards
Claus B.
--
Vista Ultimate / Windows7
Office 2007 Ultimate / 2010 Professional

On Wednesday, September 23, 2015 at 2:21:38 PM UTC-4, BottleMan wrote:
I am doing a regression on real data to find the right formula. The data is as follows:
Wall Real time
2.2 3
2.4 3.3
2.6 3.6
2.8 3.9
3 4.2
3.2 4.5
3.4 4.9
3.6 5.35
3.8 5.8
4 6.35
4.2 6.8
Excel comes up with a formula: y= 0.0153x2+0.1931x+2.8388, The R2 is 0.9991. When I compute my data with the formula these are the resulting figures:
Calc. time
3.37494
3.43472
3.49634
3.5598
3.6251
3.69224
3.76122
3.83204
3.9047
3.9792
4.05554
Not at all like the real data. I tried other trendlines but none works. What am I doing wrong? I did this with some other data and it worked fine.

Thanks for your help.

Ottmar

The scatter plot did the job. Thanks a million.

Ottmar