The "semi-exponential" generating function for these is
sumin=0nftysumnk=0F(k,n)znukovern!=exp((u−1)z)over1−zsumin=0nftysumnk=0F(k,n)znukovern!=exp((u−1)z)over1−z
which follows from the exponential formula.
These numbers are apparently called the rencontres numbers although I'm not sure how standard that name is.
Now, how do we get a formula for these numbers out of this? First note that
exp((u−1)z)=1+(u−1)z+(u−1)2over2!z2+(u−1)3over3!z3+cdotsexp((u−1)z)=1+(u−1)z+(u−1)2over2!z2+(u−1)3over3!z3+cdots
and therefore the "coefficient" (actually a polynomial in uu) of znzn in exp((u−1)z)/(1−z)exp((u−1)z)/(1−z) is
Pn(u)=1+(u−1)+(u−1)2over2!+cdots+(u−1)novern!=sumnj=0(u−1)joverj!Pn(u)=1+(u−1)+(u−1)2over2!+cdots+(u−1)novern!=sumnj=0(u−1)joverj!
since division of a generating function by 1−z1−z has the effect of taking partial sums of the coefficients.
The coefficient of ukuk in Pn(u)Pn(u) (which I'll denote [uk]Pn(u)[uk]Pn(u), where [uk][uk] denotes taking the ukuk-coefficient) is then
[uk]Pn(u)=sumnj=0[uk](u−1)joverj![uk]Pn(u)=sumnj=0[uk](u−1)joverj!
But we only need to do the sum for j=k,ldots,nj=k,ldots,n; the lower terms are zero, since they are the ukuk-coefficient of a polynomial of degree less than kk. So
[uk]Pn(u)=sumnj=k[uk](u−1)joverj![uk]Pn(u)=sumnj=k[uk](u−1)joverj!
and by the binomial theorem,
[uk]Pn(u)=sumnj=k(−1)j−koverk!(j−k)![uk]Pn(u)=sumnj=k(−1)j−koverk!(j−k)!
Finally, F(k,n)=n![uk]Pn(u)F(k,n)=n![uk]Pn(u), and so we have
F(k,n)=n!sumnj=k(−1)j−koverk!(j−k)!F(k,n)=n!sumnj=k(−1)j−koverk!(j−k)!
No comments:
Post a Comment