Mean and standard deviation on the HP25

[Dieter]

Hi all,

I would like to ask the HP25 owners here to do a little test that might show how this calculator behaves different from later devices.

First of all the Σ+ function stores Σx, Σy, Σx², Σxy and n, but not Σy². So the standard deviation of y cannot be calculated, only s_x is calculated. This also applies to the mean, only the mean of x is returned.

But... newer calculators return mean and standard deviation for both X and Y, and they overwrite the respective stack registers. If the stack is filled with 1, 2, 3 and 4, this is what an HP67 or HP41 returns:

Code:

mean:
T 1      1
Z 2  =>  2
Y 3      y-mean
X 4      x-mean

standard deviation:
T 1      1
Z 2  =>  2
Y 3      y-sdev
X 4      x-sdev

This seems to be different on the HP25:

Code:

mean:
T 1      2
Z 2  =>  3
Y 3      4
X 4      x-mean

standard deviation:
T 1      2
Z 2  =>  3
Y 3      4
X 4      x-sdev

So the HP25 does not seem to overwrite X, instead it calculates the respective value and pushes the stack.

Could the HP25 owners here please try and confirm this?
Or correct me if I'm wrong, of course. ;-)

Dieter

[PANAMATIK]

(12-01-2017 06:16 PM)Dieter Wrote:  Could the HP25 owners here please try and confirm this?

Dieter

I confirm this behavior of the HP-25. Even if number entry is in progress, the stack is pushed and either mean or sdev is placed in the X register. I always assumed, that the reason why Σy² is not calculated, is the only 8 registers available, and using Σ+ within a program would only leave two registers for general use. But also the fully used 2k program memory could be the reason for only calculating x mean.

Bernhard

[Dieter]

(12-01-2017 06:35 PM)PANAMATIK Wrote:  I confirm this behavior of the HP-25. Even if number entry is in progress, the stack is pushed and either mean or sdev is placed in the X register.

Thank you very much for your feedback. I have checked this behaviour on your HP25 emulator which, I assume, runs the original firmware and thus should behave like the real thing. The reason why I'm asking is another HP25 simulator that does it differently. Which seems to get fixed now. ;-)

While we're at it: there is one thing that's a bit strange on your emulator. If you press a key on the PC keyboard that does not have an assignment, the last valid key seems to get repeated. Example: Type 123, then "A" or "C" or "Esc" several times. The "3" is repeated.

This also happens with the "Alt" key, so that common key sequences like Alt+Tab for switching to another program may change the displayed result. Example: calculate log 7 and switch to your word processor to insert the result – as soon as the Alt-key is pressed the display changes to "7". Maybe this can be fixed some time. ;-)

Dieter