A problem with important applications in stock market analysis and music information retrieval is order-preserving matching. This problem is a recently introduced variant of the string matching problem that searches for substrings in the text whose natural representation matches the natural representation of the pattern. The natural representation of a string X is a string that contains the rankings of the characters occurring at each position of X. Then, order-preserving matching regards the internal structure of the strings rather than their absolute values. But both stock market analysis and music information retrieval require more flexibility: not only the substrings with exactly the same structure are of interest, but also the ones that are similar. In this paper, we propose an approximate version of order-preserving matching based on the δγ- distances that permit an individual error between the ranking of corresponding symbols (bounded by δ) and a global error of all the positions (bounded by γ). We present an algorithm that solves this problem in O(nm+m log m). Experimental results verify the efficiency of the proposed algorithm.