sequential similarity of similar scale.

I want a gui in which I select a portion of a price history and have the computer disect it into equal parts specified by the user. Then I would like the first portion scale( x axis)d by a range specified by the user in 0.001 increments and a correlation test run on the entire price history( at steps specified by the user) to isolate periods of high correlation with the scaled portion of the price history, then l would like the second portion tested with the sequential data from the results of the first list, one of the top 5 correlation with the most similar scale to that of the first test is used( the scale similarity range is specified by the user). Then the second test's scale is used as the scaling [login to view URL] the following test is conducted with sequential portion of the price history after test 2. one of the top 5 correlations with a similar scale is pushed forward to be tested with the fourth portion. etc.