in random steps, in blue. After interpolation and resampling (with XlXtrFun and
good setting) the green line with 10min steps, used for comparison with other
|
| Fig.1, One of our problems, meteo temperature acquired by a station practically in random steps, in blue. After interpolation and resampling (with XlXtrFun and good setting) the green line with 10min steps, used for comparison with other |
Interpolation, Downsampling, Oversampling, Fitting, Derivate, Resampling, are useful tools to analyse or rescaling a data set coming from a time series or from a continuous measures. This is NOT a course on the mathematics subtended only a suggestions to use one old software very interesting.
Above, in Fig.1 one of our problem, a data series acquired without a systematic step. Practically we can think to a noise in the X axis, we have 140 and more measure point every day but the space between is not 10 minutes. (be careful with so large noise in X we can not use a Pearson correlation, nor OLS, nor ILS).
Of course there are more for Matlab/Octave and for "R" devoted to solve this problem. But it is still difficult for my students to grasp all parameters used in those routines and often the subtended errors.
Accounting of all I said previously, I have selected Excel to better calculate the derivate. Yes, Excel another time, see the software section.
First one; under this link there is the original web page of XlXtrFun, read accurately and, also, read the other two pages, the SurfGen page and Download page.
From this web site we can copy a good description of XlXtrFun:
Three Dimensional Interpolate and Extrapolate Functions; Two Dimensional Interpolate, Extrapolate, Cubic Spline, Polynomial Curve Fitting, Line Intersections, Minima, Maxima, First Derivative, Second Derivative, and other functions.
XlXtrFun.xll is a collection of functions which extends the capabilities of Microsoft Excel; developed primarily to facilitate, interpolation of 2-dimensional and 3-dimensional data, and simplify 2-variable curve fitting. XlXtrFun has been used for years by engineering and research and development personnel on every continent who need to interpolate, extrapolate, and curve fit data rapidly, reliably, and with a virtually non-existent learning curve.
If you work with real-life data and want to interpolate, extrapolate, or curve fit, then you will find these functions very useful.
Unfortunately the download in the official page not work and I suppose next week/year the other pages disappears (but I have a copy here). Therefore, I work hard to find all versions of this software looking for in Internet for a wail, and find those:
I check with Office from version 97 to version 2003 and work well, with OS from WinNT4.0 to WinXP.
Please find XlXtrFun instructions in Internet, you can find many use and examples, and some suggestion to install it in "new" Office and OS.
Running on Win7 and Office2007
YES it work in 32 bits editions, but the installation is different and this is NOT a course on MS operating system. Looking for in Internet you can find instruction.
I have not find report to use of this software in Linux, but one ask about use on Office for MAC.
Alternatively you can find "where" your Win/Office put the .XLL files, where put the .CNT, where are the .HLP, where the .DLL and copy by hands the opportune file of XlXtrFun in those folders.
Dear students the life is short and bad, so DO NOT ASK ME on the method to install this software in your PC. If you discover something on windows, office, or better Linux, bad you can not do.
Looking for the different version I find some examples. I collect all in a single .ZIP file but, really, I do not understand how to use some files. Enclosed, for my students there is the example of Fig.1 with possibility to change parameters. In any case below there is the file.
The help and examples file is: GVisco_collection_of_XlXtrFun_examples_v1-02.ZIP, (1177289 B).
Following some suggestion for the use of Excel with large data matrix, and an alternative.
Donations if you have a Bitcoin wallet
If you have a nice fat wallet you can send a few cents to this code 1ETv1mJTZTB7EiXwZkucF2BTA512Rkaw6 maybe with a message (be careful to the transaction cost, send 0.1 and pay 0.01 is not a good idea).
1ETv1mJTZTB7EiXwZkucF2BTA512Rkaw6 <== my code , Thank you.
Donations if you have not a Bitcoin wallet
If you like this work and measures you can make a small donation of Bitcoin. You do not need a wallet to make a donation, indeed even those who have Bitcoin wallet is better to pass for a web site that "gift" currency since this does not have a transaction cost.
To make a donation you must use some of your web-time. Go to www.GreenBitCo.in, enter my code 1ETv1mJTZTB7EiXwZkucF2BTA512Rkaw6 enter the Captcha code that you see, press "Claim Now" button. Wait 15 minutes on the publicity page, when the 15 minutes pass fill the new Captcha and press "Claim Now" again. To get some value are necessary 10 pages or more. Be careful, do not use roulette, dices, casino and other stealing web site.
1ETv1mJTZTB7EiXwZkucF2BTA512Rkaw6 <== my code , Thank you.
| Universita' Degli Studi di Roma La Sapienza |
Dr. G. Visco appointed professor for chemometrics & .... |
Corso di Laurea in: Scienze Applicate ai Beni Culturali ed alla Diagnostica per la loro Conservazione |
