Deje $n$ ser un número de personas. Al menos dos de ellos pueden ser nacido en el mismo día del año con una probabilidad de: $$1-\prod_{i=0}^{n-1} \frac{365-i}{365}$$
But what is the probability that at least two of them are born on two consecutive days of the year (considering December 31st and January 1st also consecutive)? It seems a good approximation is: $$1-\prod_{i=0}^{n-1} \frac{365-2 \times i}{365}$$
Sin embargo, la simulación de pseudo-aleatorios enteros con Python, el 99%-de los intervalos de confianza pueden ser ligeramente diferentes. Así que usted tiene cerrada la leche de fórmula?
Resultados de la simulación con Python. Aquí hay un 99% de confianza de los intervalos siguientes:
Number of people: 1 Lower bound: 0.0 Upper bound: 0.0
Number of people: 2 Lower bound: 0.00528 Upper bound: 0.00567
Number of people: 3 Lower bound: 0.01591 Upper bound: 0.01657
Number of people: 4 Lower bound: 0.03185 Upper bound: 0.03277
Number of people: 5 Lower bound: 0.0528 Upper bound: 0.05397
Number of people: 6 Lower bound: 0.07819 Upper bound: 0.07959
Number of people: 7 Lower bound: 0.10844 Upper bound: 0.11006
Number of people: 8 Lower bound: 0.14183 Upper bound: 0.14364
Number of people: 9 Lower bound: 0.17887 Upper bound: 0.18086
Number of people: 10 Lower bound: 0.21816 Upper bound: 0.2203
Number of people: 11 Lower bound: 0.25956 Upper bound: 0.26183
Number of people: 12 Lower bound: 0.30306 Upper bound: 0.30544
Number of people: 13 Lower bound: 0.34678 Upper bound: 0.34925
Number of people: 14 Lower bound: 0.39144 Upper bound: 0.39397
Number of people: 15 Lower bound: 0.43633 Upper bound: 0.4389
Number of people: 16 Lower bound: 0.48072 Upper bound: 0.48331
Number of people: 17 Lower bound: 0.52476 Upper bound: 0.52734
Doy aquí algunos de los resultados con un ajustado aproximación de la fórmula, el uso de Wolfram Alpha: $$\left( 1 - \frac{n-1}{2 \times 365 + n-1} \right) \times \left( 1-\prod_{i=0}^{n-1} \frac{365-2 \times i}{365} \right)$$
However, this is just a tweak, ans is clearly wrong for $n=33$ desde:
Number of people: 33 My guess: 0.91407
Number of people: 33 Lower bound: 0.94328 Upper bound: 0.94447
Gracias a Jacopo Notarstefano, leonbloy, y Morón, aquí está la (correcta) fórmula: $$ 1-\sum_{k=1}^{n}\frac{1}{365^{n-k}k}\left(\prod_{i=1}^{k-1}\frac{365-\left(k+i\right)}{365\times i}\right)\sum_{j=0}^{k-1}\left(-1\right)^{j}C_{k}^{j}\left(k-j\right)^{n} $$
Y aquí están los resultados de los cálculos utilizando esta fórmula con la de Python:
Number of people: 1 Probability: 0.0
Number of people: 2 Probability: 0.005479452
Number of people: 3 Probability: 0.016348283
Number of people: 4 Probability: 0.032428609
Number of people: 5 Probability: 0.053459591
Number of people: 6 Probability: 0.079104502
Number of people: 7 Probability: 0.108959718
Number of people: 8 Probability: 0.14256532
Number of people: 9 Probability: 0.179416899
Number of people: 10 Probability: 0.218978144
Number of people: 11 Probability: 0.260693782
Number of people: 12 Probability: 0.304002428
Number of people: 13 Probability: 0.34834893
Number of people: 14 Probability: 0.393195856
Number of people: 15 Probability: 0.438033789
Number of people: 16 Probability: 0.482390182
Number of people: 17 Probability: 0.525836596
Number of people: 18 Probability: 0.567994209
Number of people: 19 Probability: 0.608537602
Number of people: 20 Probability: 0.647196551
Number of people: 21 Probability: 0.683756966
Number of people: 22 Probability: 0.718059191
Number of people: 23 Probability: 0.749995532
Number of people: 24 Probability: 0.779509664
Number of people: 25 Probability: 0.806569056
Number of people: 26 Probability: 0.831211564
Number of people: 27 Probability: 0.853561895
Number of people: 28 Probability: 0.873571839
Number of people: 29 Probability: 0.892014392
Number of people: 30 Probability: 0.906106867
Number of people: 31 Probability: 0.919063161
Number of people: 32 Probability: 0.928791992
Number of people: 33 Probability: 0.944659069