Calculator site update - 2005/05/16 - Printable Version +- HP Forums (https://archived.hpcalc.org/museumforum) +-- Forum: HP Museum Forums (https://archived.hpcalc.org/museumforum/forum-1.html) +--- Forum: Old HP Forum Archives (https://archived.hpcalc.org/museumforum/forum-2.html) +--- Thread: Calculator site update - 2005/05/16 (/thread-73527.html) |
Calculator site update - 2005/05/16 - Valentin Albillo - 05-16-2005 Hi, all:
I've just updated my still fledgling (time, time!) calculator-related site, which you can access using All of them do include a program, full documentation and examples, and as of now you can find the following articles there:
Hope they make interesting reading. I'll keep updating my site with new articles in the near future, as time permits.
Best regards from V.
Re: Calculator site update - 2005/05/16 - Namir - 05-16-2005 Cool articles Valentin. I really like the one for the HP41C and how to use the Advantage ROM to perform polynomial curve fitting. I agree with your comments on that ROM --- wish HP had introduced it earlier. How about a multiple regression program using the Advatnate ROM? Regards,
Namir
Re: Calculator site update - 2005/05/16 - Valentin Albillo - 05-17-2005 Hi, Namir: Namir wrote: "Cool articles Valentin [...] How about a multiple regression program using the Advantage ROM?"
Thanks for your interest and kind comments, Namir. As for your question, if by "multiple regression" you mean a *linear* regression in several variables, i.e., to fit: y = a1.x1+a2.x2+ ... + an.xnto a set of data, in such a way as to minimize the error by some criterium (say, least squares) then the mathematical treatment is nearly identical to the polynomial regression case, and so an extremely simple modification to a polynomial regression program would do. However, the program featured in my article does not compute a polynomial regression but a polynomial *fit*, i.e., there's no error to minimize, the resulting polynomial is an exact fit and actually passes through all the points given. That said, it's a very simple affair to modify the program to make it compute a least squares polynomial regression instead, and then to further modify it to compute a multi-variable linear regression. But the article was just a commemorative one for the Advantage ROM and I wanted to keep things simple. Should I actually need to perform that kind of computation using one of my handhelds, it would be preferable to write a program for the 42S, which is faster, has more RAM and a larger display, plus it's got convenient menus for data entry and results output. And in the case of multiple linear regression, my beloved bare-bones HP-71B includes statistic statements that can do it off-the-shelf with no programming involved. :-)
Best regards from V .
Re: Calculator site update - 2005/05/16 - Earl Kubaskie - 05-18-2005 I read a short story some years back (Omni Magazine, January 1979) with the same basis as your "Time Traveller", but a different ending. In Newton's last 34 years, he turned from the scientific brilliance of his youth toward persuits of religion and alchemy. In an effort to prevent Newton's "collapse" (actually caused by a temporary mental disorder), it was decided to travel back in time to give Newton the tools to continue his masterworks - a calculator. Upon presenting the calculator to Sir Isaac, the traveller demonstrated a calculation sequence that gave a result of 666, causing Sir Isaac to become nearly deranged, and to eject the traveller from his home as a servant of Satan - after which Newton abandoned science in favor of religion and alchemy.
|