what's wrong with 'smooth' and 'find' commands in this code?

x = [0 0 0 1 1 1 0 1 0 0 1 1 1 0 0 0 0];

y = smooth(x,5);

find(y(:) == 0.6)

ans = Empty matrix: 0-by-1

Running above script gives empty matrix though definitely 0.6 elements exit. What have I done wrong?

Answer 1

[Grzegorz Knor]

Look at result of this line:

y- round(y*10)/10

In vector y there is no exact 0.6 value.

Try this code:

x = [0 0 0 1 1 1 0 1 0 0 1 1 1 0 0 0 0];
y = smooth(x,5);
find(abs(y(:)-0.6)<eps)

[Taehwan]

Thank you for the comment and the results. One thing I still can't catch is that why 'smooth' doesn't give exact 0.6 (= 3/5) answer... while others are exact 0.2, 0.4, 0.8, etc.

[Grzegorz Knor]

It is 0.6, because difference between result and exact value is 1.11022302462516e-16 (on my PC) and it is smaller than floating-point relative accuracy 2.22044604925031e-16 (on my PC).

Compare with:

format long

x = single([0 0 0 1 1 1 0 1 0 0 1 1 1 0 0 0 0])

y = smooth(x,5)

Answer 2

[Jan]

This is one of the mayn effects of the limited precision of floating point values. See: Answers: Tag=faq6.1.

[Taehwan]

Thank both of you..