What Alternatives Exist for Fitting Models to Erratic Data in Excel?

[HalcyonStorm]

Homework Statement

I need to fit a model to some data, where y = systolic blood pressure and x = time in weeks. The problem is, all of the 'usua' trendline options on Excel produce awful R squared values. Is there some other method I can do to fit a different sort of model that would be accurate?

Homework Equations

None that I know.

The Attempt at a Solution

Only thing I can put here are the graphs, but that seems a little pointless.

Weeks (x)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Systolic BP (mmHg) (y)
135
115
130
110
120
125
130
130
115
125
120
130
140
115
125
120

Thanks heaps!

 

[scurty]

I'm not sure anything other than a linear fit will be good. That data, from what I saw when I plotted it, is pretty scattered. That's why your R squared value is so low. You will either need more data or more precise data in order to get a better trendline.

 

[Ray Vickson]

Homework Statement

I need to fit a model to some data, where y = systolic blood pressure and x = time in weeks. The problem is, all of the 'usua' trendline options on Excel produce awful R squared values. Is there some other method I can do to fit a different sort of model that would be accurate?

Homework Equations

None that I know.

The Attempt at a Solution

Only thing I can put here are the graphs, but that seems a little pointless.

Weeks (x)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Systolic BP (mmHg) (y)
135
115
130
110
120
125
130
130
115
125
120
130
140
115
125
120

Thanks heaps!

When plotted, your data looks almost random, with considerable "noise" masking the signal. That is why your R is so large---as it should be! It would be a great mistake to try to fit an accurate formula to random data.

RGV