Indexing functions

[FDS]

First of all a happy prosperous and especially healthy 2019 to all! I have apparently a misunderstanding when to use indexing versus vectorisation. In the attached worksheet indexing works but vectorisation doesn't. Could someone help me out with this because I seem to be lost.

Best regards, Filip

[terryhendicott]

Hi,

In the first instance of indexing, f(p,i) returns a single value at any point p and thus can be numerically integrated with respect to "p".

In the second instance of vectorisation, f(p) returns a vector as the result. "p" as an input can be either a single value or a vector of similar size to PA50. It is not possible to numerically integrate a function that returns a vector at every point. The integration thus fails.

Cheers

Terry

[Werner_E · Accepted solution]

When doing calculations where you later want to fill in vectors of data I would strongly recommend to set up all those calculations as functions dealing with scalars only and later call those functions vectorized with the data vectors as arguments.

See attached worksheet.

[FDS]

Terry, thank you so much for the clarification as this explains the why.

Werner, thank you for your additional hint. It is indeed a much clearer way to do it the way you propose. I will try to use this not only for this worksheet but also for future ones.

All the best, Filip