Post Reply 
Little explorations with HP calculators (no Prime)
04-03-2017, 06:50 PM
Post: #112
RE: Little explorations with the HP calculators
(04-03-2017 05:05 PM)pier4r Wrote:  All fine, thanks for sharing the methof of best fit (I just asked the 50g).

I've got limited experience with fitting curves, but have used Excel several times to do it so I tend to gravitate in that direction. I like being able to quickly tweak the x and y values to see how I can manipulate things for a tighter fit. I didn't spend a lot of time on this one (as you can tell Smile ).

Using the same approach, I took another look at the table and came up with a different formula (using the proper column of data this time) for the curProbV values <3826 which is an even tighter fit (all values from the function are within 0.5 of the table): \[variance=(6.8826\cdot curProbV^{0.5025})-435\]

Though not quite as tight (max delta 1.08), this formula is fairly close for the >3825 inputs: \[variance=416-(7.426\cdot(7230-curProbV)^{0.4953})\]

Despite the closeness in fitting (\(R^2\) is 0.9999 or better in both cases), the question still remains as to whether rounding and/or ceiling the results will provide a useful replacement for the table lookups. Hopefully these (or something like them) will give you what you need!
Find all posts by this user
Quote this message in a reply
Post Reply 


Messages In This Thread
RE: Little explorations with the HP calculators - DavidM - 04-03-2017 06:50 PM



User(s) browsing this thread: 2 Guest(s)