Engineer's Book

Engineers – Telecoms – Physics – Teaching – Innovations

Khi2 Tests – 04 – Probabilities Densities – Mathematics or Engineering ?

Written By: Jean-Paul Cipria - Mar• 10•17
Khi2 - Densities and Probabilities - ©J.P. Cipria - 10/03/2017

Khi2 – Densities and Probabilities – Ref. [CIPRIA-2017]

How Agile, Scrum, Programs Developpements can be more Innovative ? For example how to display Khi² Densities ? How to calculate Right Probabilities with just Simpson Algorithm ? Why we can do with this Development Method ?

.

Created :2017-03-10 12:46:01. – Modified : 2017-07-19 10:25:47.


Difficult ! Master Level.

Difficult ! Master Level.


<strong><span style="color: #ff0000;">Expériences En Construction</span></strong>

Expériences En Construction.

Do we use Mathematics or Physicists Equations ?

Engineers, do you understand what you write ?

Ref. [CIPRIA-2107] : Are methods to use equations in algorithm simplify or not understanding ? And how understanding  help engineering, scrum or agile development in high level technologies ?

Evidences ?

Matlab Program Example First Step DEMINING :

% Khi² Probability.
% -------------------------------
% Seemingly Simple Equations with NOT Understanding parameters :
% ---------------------------------------------
Khi2 = power(b,-a).*power(x,a-1).*exp(-x/b)./gamma(a);

End of Matlab Example

Do we use K. Pearson Equations then ?

First Part : do we understand khi² physical phenomenon ?

Physical phenomenon for khi² is the … probability or (repartition function for mathematician).

Khi2 - Results by K. Pearson – ©J.P. Cipria - 2017.

Khi2 – Results by K. Pearson  [PEARSON-1900] – © JP Cipria – 2017.

Second part : do we compare with K. Pearson Tabulation Values ?

We compare for K. Pearson n^{\prime}=7 and \chi^2=6 then Probabilities(\chi^2>P)=0.423190 at your left hand in the following drawing and we read for the same values at you right hand with Matlab plotting

Kearson and Cipria Khi² Probabilities Comparisons

Kearson and Khi² Probabilities Comparisons – © JP Cipria – 2017.

Third Part : we replace all mathematician equations by physicians ones ?

Do we compare Mathematics and Physics Algorithms ?

Do we use rules for High Technical Scientific Algorithms ?

Simpliest functions ?

  • Gamma function ? Don’t use it. You don’t know anything on this « codage » function.
  • FFT, filter, delay-group functions ? Don’t use it. If you are an engineer then they don’t « agree » with your expectations.

Object Programmation ?

  • Object Programmation : Don’t use, DON’T use, DON’T USE ! Use O.P. for Sciences is a fault. Then you are all in fault 😉 ! Heavy, Complex, Difficult, Inadapted for 70% of Engineer’s Minds Types (See HP), Need three days to display « Hello » : Exemple with Android … long list …
LA ZApplication - Android-Screenshot-Application

LA ZApplication – Android-Screenshot

 

  • Mouse-over without solicited customer action : Don’t …
  • Simulink : Use with extremely precautionsssssssss … but … don’t use in a first time !

Others booooh functions ?

  • Goto : …. Don’t …
  • Loop without a watchdog : Don’t …

.

… Following … 10/03/2017

.

References

  • [PEARSON-1900] : PEARSON, Karl. « On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. » Philos. Magazine, 5th Series 50 157–175., Reprinted in Karl Pearson’s Early Statistical Papers (1948) 339–357 Cambridge Univ. Press., 1900.
    .
  • [CIPRIA-2017] : CIPRIA, Jean-Paul – « Parts of PhD Annexes . » – 2017.
    • Matlab files :
      • Khi² Probabilities and Densities Algorithms : khi2_Gamma_2017_03_10_2.m
      • Khi² Model/Measurements Fittings : Khi2_Exponentielle_2017_02_24.m
        .

.

Jean-Paul Cipria
10/03/2017

You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

Laisser un commentaire

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Anti-Spam Quiz: