COUNTIF Problem
 

We collect data every 10 minutes continuously for a month at a time. When a
set trigger value is exceeded, a sample is taken. However, only one sample
can be taken in a 24-hour period (beginning and ending at 2:00 pm). The
information I need is . . . . how many samples would be taken in 1 month at
varying trigger levels (0.75, 0.80, 0.85, etc.)? Ultimately, we would like
to know what trigger level would produce 5 samples per month.

The spreadsheet has the following column headers:

Date Time Value


Any suggestions would be greatly appreciated. If more info in needed,
please let me know.

Thanks!!!!!!!!

some of it depends on your criteria for selecting the sample. If a single
value greater than a set value is enough to trigger a sample, you would want
to find the max value during your day(2 to 2) and look at the distribution of
this max value.

I would first look at all of the data to see what probability function would
best describe the data close to the trigger point.

If the data is gaussian a three sigma level would give you about 5 triggers
a month.
and other numbers could be predicted. If it is not guassian, you will have
to play with the probabilities to get what you want.
If it takes several contiguous readings greater than your trigger you would
need to program that into your calculations.

"Eaglered" wrote:

We collect data every 10 minutes continuously for a month at a time. When a
set trigger value is exceeded, a sample is taken. However, only one sample
can be taken in a 24-hour period (beginning and ending at 2:00 pm). The
information I need is . . . . how many samples would be taken in 1 month at
varying trigger levels (0.75, 0.80, 0.85, etc.)? Ultimately, we would like
to know what trigger level would produce 5 samples per month.

The spreadsheet has the following column headers:

Date Time Value


Any suggestions would be greatly appreciated. If more info in needed,
please let me know.

Thanks!!!!!!!!