0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 120543840 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 10628640 | 6668781166080 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1026576 | 628308907776 | 303443622431870976 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 109584 | 7245893376 | 381495483224064 | 17810567950611972096 |
0 | 0 | 0 | 0 | 0 | 0 | 1 | 13068 | 104587344 | 673781602752 | 3878864920694016 | 21006340945438768128 |
0 | 0 | 0 | 0 | 0 | 1 | 1764 | 1942416 | 1744835904 | 1413470290176 | 1083688832185344 | 806595068762689536 |
0 | 0 | 0 | 0 | 1 | 274 | 48076 | 6998824 | 929081776 | 117550462624 | 14500866102976 | 1765130436471424 |
0 | 0 | 0 | 1 | 50 | 1660 | 46760 | 1217776 | 30480800 | 747497920 | 18139003520 | 437786795776 |
0 | 0 | 1 | 11 | 85 | 575 | 3661 | 22631 | 13745 | 833375 | 5019421 | 30174551 |
0 | 1 | 3 | 7 | 15 | 31 | 63 | 127 | 255 | 511 | 1023 | 2047 |
1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
This differs from the table in [DKB] only by the correction of an obvious typo. We note that the non-zero entries in row k (from the bottom) have as rational generating function
k _________ | | | | 1 | | _________________ | | | | k! | | 1 - ______ z | | | | p p=1
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.000000002 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.000000025 | 0.000000076 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.00000028 | 0.0000008 | 0.0000014 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.000003 | 0.0000078 | 0.000013 | 0.0000175 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.000025 | 0.000067 | 0.00011 | 0.000144 | 0.000167 |
0 | 0 | 0 | 0 | 0 | 0 | 0.000198 | 0.000514 | 0.000827 | 0.00104 | 0.00119 | 0.00128 |
0 | 0 | 0 | 0 | 0 | 0.00139 | 0.00340 | 0.00520 | 0.00649 | 0.007305 | 0.00778 | 0.00804 |
0 | 0 | 0 | 0 | 0.00833 | 0.01903 | 0.027822 | 0.03375 | 0.03734 | 0.03937 | 0.04047 | 0.04105 |
0 | 0 | 0 | 0.04167 | 0.08681 | 0.12008 | 0.14094 | 0.15294 | 0.15950 | 0.16298 | 0.16479 | 0.16571 |
0 | 0 | 0.16667 | 0.30556 | 0.39352 | 0.44367 | 0.47081 | 0.48506 | 0.49242 | 0.49617 | 0.49807 | 0.49903 |
0 | 0.50000 | 0.75000 | 0.87500 | 0.93750 | 0.96875 | 0.98438 | 0.99219 | 0.99609 | 0.99805 | 0.99902 | 0.99951 |
1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
This table is perhaps more fun to read than the more extensive Table 4.5.2 in [DKB]. In the context of the Pez problem we see, of course, that for N distinct flavors, the probability that there is a flavor missed by the sharer (or victim!) goes to zero as the number of dispensers increases (trace the probabilities up a diagonal), while if we consider the number of dispensers fixed at k, and the number of flavors (hence the height of the dispensers) to increase, the probability of missing a flavor [(k-1)!*g(m,k)] seems to approach 1 (This limit is easily verified using the iterated difference formula for g(m,k)).