Knowledge (XXG)

Banach's matchbox problem

Source 📝

229: 34:
Suppose a mathematician carries two matchboxes at all times: one in his left pocket and one in his right. Each time he needs a match, he is equally likely to take it from either pocket. Suppose he reaches into his pocket and discovers for the first time that the box picked is empty. If it is assumed
629: 343: 674: 219: 185: 133: 103:
be the number of matches removed from this one before the left one is found to be empty. When the left pocket is found to be empty, the man has chosen that pocket
704: 416: 724: 436: 388: 153: 101: 73: 53: 735: 444: 243: 754:
Feller, William, An Introduction to Probability Theory And Its Applications, Third Edition, Wiley, 1968, Chapter VI, section 8
83:
Without loss of generality consider the case where the matchbox in his right pocket has an unlimited number of matches and let
222: 27:. Feller says that the problem was inspired by a humorous reference to Banach's smoking habit in a speech honouring him by 796: 791: 677: 349:
Returning to the original problem, we see that the probability that the left pocket is found to be empty first is
638: 776: 190: 158: 106: 683: 393: 709: 421: 352: 138: 86: 58: 38: 28: 785: 24: 20: 228: 624:{\displaystyle P=P=2P={\binom {2N-k}{N-k}}\left({\frac {1}{2}}\right)^{2N-k}} 31:, but that it was not Banach who set the problem or provided an answer. 338:{\displaystyle P={\binom {N+m}{m}}\left({\frac {1}{2}}\right)^{N+1+m}} 706:
matches, the expected number of matches in the second box is
418:
because both are equally likely. We see that the number
55:
matches, what is the probability that there are exactly
712: 686: 641: 635:
The expectation of the distribution is approximately
447: 424: 396: 355: 246: 193: 161: 141: 109: 89: 61: 41: 718: 698: 668: 623: 430: 410: 382: 337: 213: 179: 147: 127: 95: 67: 47: 581: 549: 292: 271: 35:that each of the matchboxes originally contained 8: 438:of matches remaining in the other pocket is 711: 685: 650: 645: 640: 606: 592: 580: 548: 546: 490: 446: 423: 400: 395: 354: 317: 303: 291: 270: 268: 245: 203: 192: 160: 140: 108: 88: 60: 40: 736:List of things named after Stefan Banach 227: 747: 236:matches remaining in the other pocket. 232:Distribution of probability of having 7: 669:{\displaystyle 2{\sqrt {N/\pi }}-1} 553: 275: 187:failures in Bernoulli trials with 155:is the number of successes before 14: 680:.) So starting with boxes with 540: 522: 510: 491: 472: 463: 451: 377: 359: 262: 250: 223:negative binomial distribution 174: 162: 122: 110: 1: 813: 75:matches in the other box? 678:Stirling's approximation 19:is a classic problem in 676:. (This is shown using 720: 700: 670: 625: 432: 412: 384: 339: 237: 215: 181: 149: 129: 97: 69: 49: 17:Banach's match problem 721: 701: 671: 626: 433: 413: 385: 340: 231: 216: 214:{\displaystyle p=1/2} 182: 180:{\displaystyle (N+1)} 150: 130: 128:{\displaystyle (N+1)} 98: 70: 50: 797:Probability problems 710: 699:{\displaystyle N=40} 684: 639: 445: 422: 394: 353: 244: 191: 159: 139: 107: 87: 59: 39: 792:Applied probability 411:{\displaystyle 1/2} 716: 696: 666: 621: 428: 408: 380: 335: 238: 211: 177: 145: 125: 93: 65: 45: 763:Feller, page 238. 719:{\displaystyle 6} 658: 600: 579: 431:{\displaystyle K} 383:{\displaystyle P} 311: 290: 148:{\displaystyle M} 96:{\displaystyle M} 68:{\displaystyle k} 48:{\displaystyle N} 804: 764: 761: 755: 752: 725: 723: 722: 717: 705: 703: 702: 697: 675: 673: 672: 667: 659: 654: 646: 630: 628: 627: 622: 620: 619: 605: 601: 593: 586: 585: 584: 578: 567: 552: 494: 437: 435: 434: 429: 417: 415: 414: 409: 404: 389: 387: 386: 381: 344: 342: 341: 336: 334: 333: 316: 312: 304: 297: 296: 295: 286: 274: 221:, which has the 220: 218: 217: 212: 207: 186: 184: 183: 178: 154: 152: 151: 146: 134: 132: 131: 126: 102: 100: 99: 94: 74: 72: 71: 66: 54: 52: 51: 46: 812: 811: 807: 806: 805: 803: 802: 801: 782: 781: 773: 768: 767: 762: 758: 753: 749: 744: 732: 708: 707: 682: 681: 637: 636: 588: 587: 568: 554: 547: 443: 442: 420: 419: 392: 391: 351: 350: 299: 298: 276: 269: 242: 241: 189: 188: 157: 156: 137: 136: 105: 104: 85: 84: 81: 57: 56: 37: 36: 12: 11: 5: 810: 808: 800: 799: 794: 784: 783: 780: 779: 772: 771:External links 769: 766: 765: 756: 746: 745: 743: 740: 739: 738: 731: 728: 715: 695: 692: 689: 665: 662: 657: 653: 649: 644: 633: 632: 618: 615: 612: 609: 604: 599: 596: 591: 583: 577: 574: 571: 566: 563: 560: 557: 551: 545: 542: 539: 536: 533: 530: 527: 524: 521: 518: 515: 512: 509: 506: 503: 500: 497: 493: 489: 486: 483: 480: 477: 474: 471: 468: 465: 462: 459: 456: 453: 450: 427: 407: 403: 399: 379: 376: 373: 370: 367: 364: 361: 358: 347: 346: 332: 329: 326: 323: 320: 315: 310: 307: 302: 294: 289: 285: 282: 279: 273: 267: 264: 261: 258: 255: 252: 249: 210: 206: 202: 199: 196: 176: 173: 170: 167: 164: 144: 124: 121: 118: 115: 112: 92: 80: 77: 64: 44: 29:Hugo Steinhaus 23:attributed to 13: 10: 9: 6: 4: 3: 2: 809: 798: 795: 793: 790: 789: 787: 778: 775: 774: 770: 760: 757: 751: 748: 741: 737: 734: 733: 729: 727: 713: 693: 690: 687: 679: 663: 660: 655: 651: 647: 642: 616: 613: 610: 607: 602: 597: 594: 589: 575: 572: 569: 564: 561: 558: 555: 543: 537: 534: 531: 528: 525: 519: 516: 513: 507: 504: 501: 498: 495: 487: 484: 481: 478: 475: 469: 466: 460: 457: 454: 448: 441: 440: 439: 425: 405: 401: 397: 390:which equals 374: 371: 368: 365: 362: 356: 330: 327: 324: 321: 318: 313: 308: 305: 300: 287: 283: 280: 277: 265: 259: 256: 253: 247: 240: 239: 235: 230: 226: 224: 208: 204: 200: 197: 194: 171: 168: 165: 142: 135:times. Then 119: 116: 113: 90: 78: 76: 62: 42: 32: 30: 26: 25:Stefan Banach 22: 18: 759: 750: 634: 348: 233: 82: 33: 16: 15: 777:Java applet 21:probability 786:Categories 742:References 661:− 656:π 614:− 573:− 562:− 535:− 485:− 225:and thus 730:See also 79:Solution 499:< 366:< 788:: 726:. 694:40 714:6 691:= 688:N 664:1 652:/ 648:N 643:2 631:. 617:k 611:N 608:2 603:) 598:2 595:1 590:( 582:) 576:k 570:N 565:k 559:N 556:2 550:( 544:= 541:] 538:k 532:N 529:= 526:M 523:[ 520:P 517:2 514:= 511:] 508:1 505:+ 502:N 496:M 492:| 488:k 482:N 479:= 476:M 473:[ 470:P 467:= 464:] 461:k 458:= 455:K 452:[ 449:P 426:K 406:2 402:/ 398:1 378:] 375:1 372:+ 369:N 363:M 360:[ 357:P 345:. 331:m 328:+ 325:1 322:+ 319:N 314:) 309:2 306:1 301:( 293:) 288:m 284:m 281:+ 278:N 272:( 266:= 263:] 260:m 257:= 254:M 251:[ 248:P 234:k 209:2 205:/ 201:1 198:= 195:p 175:) 172:1 169:+ 166:N 163:( 143:M 123:) 120:1 117:+ 114:N 111:( 91:M 63:k 43:N

Index

probability
Stefan Banach
Hugo Steinhaus
negative binomial distribution

Stirling's approximation
List of things named after Stefan Banach
Java applet
Categories
Applied probability
Probability problems

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.