Data Set Frequency

  • Thread starter Thread starter DataMan
  • Start date Start date
D

DataMan

I need to count the amount of times that a certain cycle takes place
Say a column of data starts at 2.1 and climbs up to somewhere aroun
4.2. then it shoots down to 2.4 and starts to climb back to 3.8. The
repeats. No two high numbers or low numbers are the same but a cycle o
this appoximate range has taken place.

I am trying COUNTIF but it counts every single time a number i
reached. Greater than/ Less than does not work either. Some time th
numbers will be the same as the cycle is moving so I end up gettin
bogus counts.

Example: Say column C has the following numbers in it.

2.01
2.01
2.32
2.65
3.01
3.5
3.5
3.5
3.49
3.87
3.93
2.7
2.3
2.5
2.8
3.7
4.1
3.4
3.4
3.4
3.4
2.8

With this set of numbers 2 highs where reached and then dropped off.
want to count these cycles on a much larger amount of data. (43,00
datapoints) With the highs and lows not being equal I don't know how.

Does anybody have a suggestion
 
How about doing a rolling average over N of the C data. Put this in D. Look
at the data to guess a reasonable N. Then look for minima with something
=D1=min(D:D) or Abs(d1-min(D:D)<smallvalue
OR
Fit to a sine curve; in D use =amp*SIN(B1*freq - phase)+ shift (where B has
serial numbers 1,2,3,...); Compute SumSquares of Residuals (ssrid) using
SUMXMY2() and have Solve alter the parameters amp, freq,phase and shift to
minimize ssrid. I get a fairly good fit and can more easily locate minima
in the Sine data.
 
Back
Top