RE: Any Statistics/Maths Nerds out there?
(06-30-2014, 03:04 PM)Sleepster Wrote: Solution
During the 1st session 95% of the time the readings were below 8. 95% of 3 hours is 2.85 hours.
During the 2nd session 95% of the time the readings were below 5. 95% of 4 hours is 3.8 hours.
During the 3rd session 95% of the time the readings were below 9. 95% of 2 hours is 1.9 hours.
So for 2.85 hours the readings were below 8.
For 3.8 hours the readings were below 5.
For 1.9 hours the readings were below 9.
2.85 + 3.8 + 1.9 = 8.55 hours, which is of course simply 95% of the total time of 9 hours.
lest you get swatted with a ruler again, you might wanna read this:
http://adventures-in-hosehead-land.blogs...de-to.html
RE: Any Statistics/Maths Nerds out there?
This is one of those questions where you don't have enough data to get the true answer. I think we also need to think about what we use the number for.
What do we think 95th percentile means in terms of treating the patient? To me, it's sort of the highest pressure you had, while trying to throw out short term anomalous high readings.
What if you just assume the actual pressure was equal to the 95% pressure for each time period you have a 95th percentile for? Then calculate a 95th percentile for that data. At first glance, that sort of equates to the idea of "maximum with glitches in the data thrown out."
Get the free OSCAR CPAP software
here.
Useful links.
Click
here for information on the main alternative to CPAP.
If it's midnight and a DME tells you it's dark outside, go and check it yourself.
RE: Any Statistics/Maths Nerds out there?
(06-30-2014, 07:17 PM)archangle Wrote: What if you just assume the actual pressure was equal to the 95% pressure for each time period you have a 95th percentile for? Then calculate a 95th percentile for that data.
I believe that's what the weighted average does. See Post #2.
Sleepster
INFORMATION ON APNEA BOARD FORUMS OR ON APNEABOARD.COM SHOULD NOT BE CONSIDERED AS MEDICAL ADVICE. ALWAYS SEEK THE ADVICE OF A PHYSICIAN BEFORE SEEKING TREATMENT FOR MEDICAL CONDITIONS, INCLUDING SLEEP APNEA. INFORMATION POSTED ON THE APNEA BOARD WEB SITE AND FORUMS ARE PERSONAL OPINION ONLY AND NOT NECESSARILY A STATEMENT OF FACT.
RE: Any Statistics/Maths Nerds out there?
(06-30-2014, 07:20 PM)Sleepster Wrote: (06-30-2014, 07:17 PM)archangle Wrote: What if you just assume the actual pressure was equal to the 95% pressure for each time period you have a 95th percentile for? Then calculate a 95th percentile for that data.
I believe that's what the weighted average does. See Post #2.
No. For instance, a 4 hour session with a 95 percentile of 10, and a 4 hour session of 4 would yield 10 with this method, 7 with a weighted average.
Which one is more appropriate is the question. What are we trying to indicate with a 95th percentile?
Get the free OSCAR CPAP software
here.
Useful links.
Click
here for information on the main alternative to CPAP.
If it's midnight and a DME tells you it's dark outside, go and check it yourself.
RE: Any Statistics/Maths Nerds out there? (SleepyHead)
Thanks for breaking that down to a simple example robysue. that highlighted the difference extremely well
I think the most "technically correct" answer (which was suggested over the other place) is to give a N.A. because it's not really possible to calculate. I need a lot more samples than just the 0, 50th and 95th percentiles, and max to pull a result from fitting a curve, even if I'm half kludging with a weighted average.
But it comes down to what users want to see there.
Maybe I can put a preference for the technically offended (like me) to barf abort the calculation when the data set is incomplete, or display a weighted average hack for those who don't care.
I do think it would be a bit annoying getting NA's everywhere if the card was left out for a night in the last week though :-/
I'm going to ask a sleep research Doc for advice on this topic..