Data analyzing device and method, and program for making computer execute the data analyzing method

A data analyzing method comprising the steps of:

• storing in a memory unit to store m records Ri (i=1, 2, . . . , m and m is a natural number of 2 or more) having an explanatory variable xi and an object variable yi which is a quantitative variable;
• and
• reading in a computing unit
• to read out the m records Ri from the memory unit,
• to partition the m records Ri into n small sets Gj (j=1, 2, . . . , n and n is a natural number satisfying 2.ltoreq.n.ltoreq.m),
• to calculate an average value of the object variables yi for every small set Gj,
• to rearrange the n small sets Gj in an ascending order or a descending order of the average value,
• to calculate n−1 combinations Ak in which the n rearranged small sets Gj are partitioned into two large sets which are a large set G′1k including the k small sets Gj (k=1, 2, . . . , n−1 and k is a natural number) selected in a descending order from the small set having the largest average value and a large set G′2k including the number n−k remaining small sets Gj,
• to calculate a grade of unity which is represented as the following expression for each of the n−1 combinations Ak, and
• to perform a predetermined data analyzing operation on the basis of the grade of unity:
• Grade of Unity=[[S0−(S1+S2)]/S0].times.100, where S0 is the sum of deviation squares of the object variables yi of the m records Ri,
• S1 is the sum of deviation squares of the object variables yi of the records Ri belonging to the large set G′1k, and
• S2 is the sum of deviation squares of the object variables yi of the records Ri belonging to the large set G′2k.