Comparing Absolute Deviation to Mean Absolute Deviation

Click For Summary
The discussion focuses on measuring variability in sensor data when the true value is unknown. The user seeks to compare absolute deviation and mean absolute deviation to assess precision as they modify a sensor component. They question whether a data set can be deemed accurate if all absolute deviations fall within ±5% of the mean absolute deviation. Clarification is requested on the methodology for measuring precision and the arithmetic involved in analyzing the observations. The context involves nuclear counting for medical imaging, emphasizing the need for consistent operational performance with changing components.
Nyasha
Messages
127
Reaction score
0
Hi Guys,

I am trying to measure variability in a part for my sensor which l do not know the true value. So l decided that a good way to measure variability in this case would be to measure precision of my data points as l change this part on the sensor. So l was wondering, can l compare the absolute deviation and mean absolute deviation and calculate the percentage deviation between the two, and then use the percentage deviation as a number which tells me something about variability in this part.


Thanks.
 
Physics news on Phys.org
Your description is not very clear. Please flesh it out more, perhaps with an example.
 
haruspex said:
Your description is not very clear. Please flesh it out more, perhaps with an example.

I calculated 15 absolute deviations and their mean absolute deviation. Would it be valid for me to say my data set is accurate if all the 15 absolute deviations are within ±5% of the mean absolute deviation ?
 
No, I meant a much more thorough description of what you are doing. What do you mean by "measuring the precision"? What does this involve? What, arithmetically, do you then do with the observations?
 
I am doing nuclear counting for medical imaging and l do not know the 'true' estimate of counts for my system. So to see if my system is operating consistently as l change a certain part, l am going to measure its precision. That is, as l change that part, are my number of counts within the same range.
 
The standard _A " operator" maps a Null Hypothesis Ho into a decision set { Do not reject:=1 and reject :=0}. In this sense ( HA)_A , makes no sense. Since H0, HA aren't exhaustive, can we find an alternative operator, _A' , so that ( H_A)_A' makes sense? Isn't Pearson Neyman related to this? Hope I'm making sense. Edit: I was motivated by a superficial similarity of the idea with double transposition of matrices M, with ## (M^{T})^{T}=M##, and just wanted to see if it made sense to talk...

Similar threads

  • · Replies 42 ·
2
Replies
42
Views
5K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
Replies
2
Views
2K
Replies
4
Views
1K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 2 ·
Replies
2
Views
1K
Replies
5
Views
11K
  • · Replies 28 ·
Replies
28
Views
3K