Stabilising a distribution (modal score question)
Archiver
Posts: 46,084
OK, Cool, I have the V2X working with the stamp now, thanks to Bruce
Bates for pointing me in the direction of some code in my earlier
question [noparse]:)[/noparse]
It appears that the output jitters randomly +/- one degree from the
real heading,and sometimes (but rarely) 2 degrees.
I want to stabilise the display output, whilst maintaining the most
accurate reading (yes, a degree is important here).
I tried averaging it across N=5, then N=10, and neither had any
effect. Beyond N=10, it was all too slow.
Having watched the figures, it appears that there is a clear modal
score, so I would like to work with that.
I can work out some REALLY long-winded ways to calculate a modal
score, involving a separate variable for each sample, which has
obvious drawbacks (and if n is larger then gosh!!)
Does anyone have any ideas how I can do this? I'm curious how to
calc the modal score, but if anyone has a better idea how to
stabilise the output, please let me know.
Regards,
Dwain.
Bates for pointing me in the direction of some code in my earlier
question [noparse]:)[/noparse]
It appears that the output jitters randomly +/- one degree from the
real heading,and sometimes (but rarely) 2 degrees.
I want to stabilise the display output, whilst maintaining the most
accurate reading (yes, a degree is important here).
I tried averaging it across N=5, then N=10, and neither had any
effect. Beyond N=10, it was all too slow.
Having watched the figures, it appears that there is a clear modal
score, so I would like to work with that.
I can work out some REALLY long-winded ways to calculate a modal
score, involving a separate variable for each sample, which has
obvious drawbacks (and if n is larger then gosh!!)
Does anyone have any ideas how I can do this? I'm curious how to
calc the modal score, but if anyone has a better idea how to
stabilise the output, please let me know.
Regards,
Dwain.