I have a graph with many peaks which are decreasing logarithmically and i
need to get the slope of these peaks ignoring the "Noise" which is below
these peaks.

Any suggestions

I'm sure your question makes sense to you but by definition a peak has a
slope of zero. Or do you mean the slope of a curve that includes only
the peaks? If so, plot the peak values as a 2nd series. If that is a
straight line, the trendline (or the LINEST function) will give you the
slope.

--
Regards,

Tushar Mehta
www.tushar-mehta.com
Multi-disciplinary business expertise
+ Technology skills
= Optimal solution to your business problem
Recipient Microsoft MVP award 2000-2005

In article ,
says...
I have a graph with many peaks which are decreasing logarithmically and i
need to get the slope of these peaks ignoring the "Noise" which is below
these peaks.

Any suggestions

Tushar Mehta

Sorry I probaly didnt word the question well. My graph has many peaks e.g.
20 peaks, each one smaller than the previous one. What i am trying to do is
draw a lins through the top of these peaks and then get the slope of this
line.

I have tried using the trendling but with no success. It does give me ths
slope of the whole graph which is useful but not what i am looking for.
Regards

Cathal



"Tushar Mehta" wrote:

I'm sure your question makes sense to you but by definition a peak has a
slope of zero. Or do you mean the slope of a curve that includes only
the peaks? If so, plot the peak values as a 2nd series. If that is a
straight line, the trendline (or the LINEST function) will give you the
slope.

--
Regards,

Tushar Mehta
www.tushar-mehta.com
Multi-disciplinary business expertise
+ Technology skills
= Optimal solution to your business problem
Recipient Microsoft MVP award 2000-2005

In article ,
says...
I have a graph with many peaks which are decreasing logarithmically and i
need to get the slope of these peaks ignoring the "Noise" which is below
these peaks.

Any suggestions

Is your chart an XY or line chart?
If it is a XY chart it is easier.
You will need to set up a method on the spread sheet of identifying the peaks.
the method to do this depends on the noise in the data.
if the data between max and mins is always monotonically increasing or
decreasing you could use something like
=if(and(d3d2,d3d4),d3,na())
If there is a lot of up and down movement in your curves You will need to
put in a smoothing function


another way to do it is to select on the chart points which look like they
are maxes and record the x,y posion shown next to them.
record these positions on a new section of sheet plot it on the chart and
generate your trend line

A third way is to put in an XY plot of a line with the number of peaks you
expect to have with a max X equal to the max of the chart and a min X equal
to the min of the chart. Select each point on this line individually, and
drag it to be on a peak. once you have all of them moved. delete points you
didn't use and do a trendline on this data.

"capillod" wrote:

Tushar Mehta

Sorry I probaly didnt word the question well. My graph has many peaks e.g.
20 peaks, each one smaller than the previous one. What i am trying to do is
draw a lins through the top of these peaks and then get the slope of this
line.

I have tried using the trendling but with no success. It does give me ths
slope of the whole graph which is useful but not what i am looking for.
Regards

Cathal



"Tushar Mehta" wrote:

I'm sure your question makes sense to you but by definition a peak has a
slope of zero. Or do you mean the slope of a curve that includes only
the peaks? If so, plot the peak values as a 2nd series. If that is a
straight line, the trendline (or the LINEST function) will give you the
slope.

--
Regards,

Tushar Mehta
www.tushar-mehta.com
Multi-disciplinary business expertise
+ Technology skills
= Optimal solution to your business problem
Recipient Microsoft MVP award 2000-2005

In article ,
says...
I have a graph with many peaks which are decreasing logarithmically and i
need to get the slope of these peaks ignoring the "Noise" which is below
these peaks.

Any suggestions