The easiest and fastest way would be to treat the values of X as indices, and simply define a look-up vector like this:
>> A(3:9) = [0.1,0.2,0.3,0.55,0.8,1.25,1.7];
which you can use with your X vector very simply as indices:
>> X = [3,5,4,7,8,6,5,6,7,9,5,4];
>> A(X)
ans = 0.1 0.3 0.2 0.8 1.25 0.55 0.3 0.55 0.8 1.7 0.3 0.2
Or equivalently but written more explicitly:
>> key = [3, 4, 5, 6, 7, 8, 9];
>> val = [0.1,0.2,0.3,0.55,0.8,1.25,1.7]
>> A(key) = val;
>> A(X)
This code places the values in val in the new vector A, in positions given by key. We can them simply use A(X) to get the values in A corresponding to the values/indices in X (this is called indexing, and is a very powerful part of MATLAB).
This method will work as long as all values in X are integer and are greater than zero.
Note that your example code uses parentheses to define the vector: in MATLAB, a vector like this is created by concatenating values using square brackets. Note that the square brackets are not a list operator, but a concatenation operator!
