Fibonacci

The following function generate the fibinacci series up to the sequence number m.

Description

The following function generate the fibinacci series up to the sequence number m which should be a positive integer, it may takes several minutes for big numbers of m, for a typical working station m>20 is taking considerable amount of time to be processed.

Feel free to contact us for any furthur information:
 {hasan.demirel, shahab.jafari}@emu.edu.tr
  http://faraday.ee.emu.edu.tr/shahab
  http://faraday.ee.emu.edu.tr/hdemirel
  (c) Demirel and Anbarjafari -2008

• MATLAB release: MATLAB 7.5 (R2007b)

[Anon]

Why going for recursion, when there are other ways to solve it?

Your function:
>>tic;a = fibonacci(30);toc
Elapsed time is 57.611550 seconds.
>> a

a =

      832040

David Terr's function
http://www.mathworks.com/matlabcentral/fileexchange/5093

>> tic;a = fibonacci2(30);toc
Elapsed time is 0.000082 seconds.
>> a

a =

      832040

[alain boyer]

 

[i posh]

Shahab, thanks for the example....

[Jos (10584)]

There are so many ways to arrive at a fibonacci number: see http://blogs.mathworks.com/loren/2006/05/17/fibonacci-and-filter/

This submission is not so bad in itself, although the help lacks an H1 line (used by LOOKFOR) and an example. However, as this is clearly a product of a programming exercise, I do not think this is worthwhile to post it here on the FEX. Hence my 2 stars.

[Shahab Anbarjafari]

Thanks for your comment Jos, well the reason that I post this file was some of my students want an example of using a recursive function MATLAB, so I let them have it by posting it here, not only them but maybe some other people may want to just have a review of recursive function.
I appreciate you sparing your time on commenting my programme.

[Jeff Burkey]

Shahab, cyclonic programming is the most cpu time intensive technique (i.e. slow). I would like to make the suggestion, why not teach your students multi-dimensional techniques which is what Matlab excels at best for a platform and is orders of magnitude faster. That said, working with large datasets certainly may require loops, or at best a combination of loops and dimensional math. Plus if you go parallel processing deminsonal math would be prefered. Regards, Jeff.

[Derek O'Connor]

This is a very poor example for demonstrating recursion and my lead naive students to believe that all recursion is inefficient.

 This is not true: the recursive Fibonacci generator is a classic example of using a good tool for the wrong job.

For example, fibonacci(25) makes 243,028 fibonacci() calls, all but 25 of them superfluous.

If you really need to calculate Fibonacci numbers then something such as the following function will be much faster.

function result = IFib(n);
F0 = 0;
F1 = 1;
for k = 2:n
    F2 = F1 + F0;
    F0 = F1;
    F1 = F2
end;
result = F2;

If you need to demonstrate recursion to students then choose a better example, such as Binary Search of an ordered table. Many more examples can be found in any good book on algorithms.

Derek O'Connor

[Xu Wings]

not so professional

[Kemal]

 

[Hal 9000]

poorly written

[Shahab Anbarjafari]

:)

[Kemal]

 

[V. Poor]

 

[alain boyer]

 

[Xing Yu]

 

[Robin Torres]

 

[Mustafa]

 

[Xu Wings]

 

[Jennifer Galt]

Poor. Terribly inefficient, even for a beginner.