Predicting Dataset: Compare 5 vs 25 Numbers

[lzh]

Suppose that you have 3 datasets of varying numbers from a device. Let's say one has a range from 30-50, another has a range from 40-70, and the last one has a range from 65-80. The numbers here doesn't matter, but it illustrates my point. Each dataset holds 25 numbers.

Each dataset received its numbers from a device reading signal strengths. You the the standard deviation, the max and min and mean of each dataset. If you were to run the device again, but only collecting 5 numbers this time. How can you compare those 5 numbers with the numbers from the 3 datasets to determine which dataset it should be a part of?

I hope that question was clear. Thanks for any help!

 

[Valkarie]

The simplest approach would be to compare the five numbers from the device with the mean, max, and min of each dataset. If the five numbers fall within the range of any given dataset (i.e. their maximum is less than or equal to the max of that dataset, and their minimum is greater than or equal to the min of that dataset), then it can be assumed that the five numbers belong in that dataset. If none of the datasets fit the criteria, then the five numbers could be disregarded as they don't fit any of the datasets.