For nonnegative integers , the Stirling number of the second kindcounts the set partitions of an -element finite set into exactly unlabelled blocks of a set partition. Set , for , and for . Also for . Grassl and Levin develop these counts through explicit set partitions and a recurrence relation. [ref]
https://discrete.openmathbooks.org/more/mdm/sec_adv-partitions.html. Grassl and Levin, More Discrete Mathematics via Graph Theory, section 3.1: counting partitions
Insert one distinguished element. For and , the Stirling numbers of the second kind obeyProof outline. Inspect the block of a set partition containing . If it contains another element, removing leaves a set partition into blocks of a set partition; conversely, insert into any of those blocks of a set partition. If is alone, remove its singleton set and obtain a set partition into blocks of a set partition. The cases cannot overlap, and deletion reverses each construction. This explains the factor even though the blocks of a set partition have no names. [ref]
https://discrete.openmathbooks.org/more/mdm/sec_adv-stirling.html. Grassl and Levin, More Discrete Mathematics via Graph Theory, section 3.4: Stirling recurrence
For , start with the three two-block set partitions of . Inserting into either block of a set partition gives six possibilities. The remaining possibility is , obtained by making a singleton set. Hence . The seven set partitions in the diagram comprise four with sizes and three with sizes ; this grouping by sizes independently gives .
Exact construction of all seven two-block set partitions of : six insert into an existing block of a set partition, and one creates a singleton set
. Details. Three different counting questions. The binomial coefficient selects one -element subset. The Stirling number of the second kind groups all distinct elements into nonempty unnamed blocks of a set partition. The power counts all functions to a specified -element finite set, allowing some destination labels to go unused. At , these counts are respectively , , and . Choosing two people, making two unnamed nonempty teams, and assigning each person a red or blue label are different tasks.
When the destinations have labels. For fixed finite sets of sizes , respectively, the number of surjective functions isHere every element of must be used. Each surjective function gives a set partition of into its nonempty preimages . Conversely, assign the distinct destination labels to the blocks of a set partition in ways, where is a factorial. For two labelled teams of four people this gives : of the assignments, exactly two use only one label. Dividing arbitrary function counts by fails in general because unused labels can be permuted without changing a function. For example, cannot count set partitions. The empty-to-empty function is a surjection, consistent with . Stanley gives the bijection underlying this qualification.[ref]
For example, the inclusion-exclusion principle counts the surjections from four elements to three labelled destinations as . Thus . An independent check chooses the only two-element block of a set partition in ways; the other two elements form singleton sets.
Why these numbers also occur in algebra. Classifying all functions to an -element finite set by the size of their image givesFor a fixed set partition, the falling factorial counts assignments of distinct labels to its blocks of a set partition, and is zero when . At , this reads . Replacing by yields a polynomial identity: both sides agree at every nonnegative integer. Thus the Stirling numbers of the second kind translate powers into falling factorials. Grouping by equal values turns a counting rule into an algebraic change of coordinates. Summing without destination labels instead gives the Bell number.
Articles by others on the same topic
The Stirling numbers of the second kind, denoted as \( S(n, k) \), are a set of combinatorial numbers that count the ways to partition a set of \( n \) objects into \( k \) non-empty subsets. In other words, \( S(n, k) \) gives the number of different ways to group \( n \) distinct items into \( k \) groups, where groups can have different sizes but cannot be empty.
