Develop statistical algorithm - pattern recognition, correlation of 2 big data sets, stock quotes -- 2

We need someone to develop an algorithm that would detect patterns established by the smaller data set A within a larger data set B, using statistical methods, correlation coefficients, or other means as relevant. Both data sets represent price quotes on financial instruments. You will be provided with two data sets, containing price quotes on financial instruments (stocks) with 1-minute time intervals. First data set contains current data with 1 day worth of information (Data Set A), and the second contains historic data for the past 10 years (Data Set B). The project is split and delivered in the following stages: I. Development of the main data comparison and matching procedure: 1) Identify parameters that uniquely characterize current data set (Set A) 2) Develop algorithm to search for a segment of data in Set B that contains parameters most resembling those identified in step 1), i.e. find days of historic data that are most similar to current data set. 3) Produce a list of similar data sets from historic Set B in descending degree of similarity to Set A. II. Incorporate additional filters 1) Incorporate logic of additional filtering in the algorithm that pre-processes Set B to limit scope of comparison with Set A. III. Assist our programming team by adjusting algorithm as necessary during live testing to ensure accurate results. Required experience and qualifications: 1) Extensive experience and fluency in applied statistics. 2) Solid understanding of financial markets and price action. 3) You received formal education in math and/or statistics, or have otherwise trained in both subjects. 4) Extensive experience developing algorithms. 4) Experience managing big data is especially welcome, but is not a prerequisite. 5) Good understanding of programming methods and workflows. You don't need to be a programmer yourself, but must be able to coordinate your work with the established programming team.