Knowledge (XXG)

Quantile function

Source 📝

399: 3373: 31: 2802:) is the probability density function. The forms of this equation, and its classical analysis by series and asymptotic solutions, for the cases of the normal, Student, gamma and beta distributions has been elucidated by Steinbrecher and Shaw (2008). Such solutions provide accurate benchmarks, and in the case of the Student, suitable series for live Monte Carlo use. 1597:
function. Unfortunately, this function has no closed-form representation using basic algebraic functions; as a result, approximate representations are usually used. Thorough composite rational and polynomial approximations have been given by Wichura and Acklam. Non-composite rational approximations
1857:
This has historically been one of the more intractable cases, as the presence of a parameter, ν, the degrees of freedom, makes the use of rational and other approximations awkward. Simple formulas exist when the ν = 1, 2, 4 and the problem may be reduced to the solution of a
2195: 1461:
for use in diverse types of simulation calculations. A sample from a given distribution may be obtained in principle by applying its quantile function to a sample from a uniform distribution. The demands of simulation methods, for example in modern
1072: 1453:
entry. Before the popularization of computers, it was not uncommon for books to have appendices with statistical tables sampling the quantile function. Statistical applications of quantile functions are discussed extensively by Gilchrist.
704: 2672: 1707: 2789: 2104: 1202: 2018: 1830: 3323: 929: 593: 2349: 2115: 1445:
of a given distribution. For example, they require the median and 25% and 75% quartiles as in the example above or 5%, 95%, 2.5%, 97.5% levels for other applications such as assessing the
165: 74:
function associates with a range at and below a probability input the likelihood that a random variable is realized in that range for some probability distribution. It is also called the
320: 3475: 1941: 218: 1764: 1389: 393: 1330: 1271: 2249: 1124: 806: 764: 2550:
is a quantile function. Two four-parametric quantile mixtures, the normal-polynomial quantile mixture and the Cauchy-polynomial quantile mixture, are presented by Karvanen.
2492: 2427: 2254:
In the above the "sign" function is +1 for positive arguments, −1 for negative arguments and zero at zero. It should not be confused with the trigonometric sine function.
1522:
to invert the cdf. Other methods rely on an approximation of the inverse via interpolation techniques. Further algorithms to evaluate quantile functions are given in the
970: 3316: 1593:, its quantile function for arbitrary parameters can be derived from a simple transformation of the quantile function of the standard normal distribution, known as the 2389: 852: 2519: 2454: 2548: 3438: 3309: 1419: 2926: 608: 1530:
packages. General methods to numerically compute the quantile functions for general classes of distributions can be found in the following libraries:
2576: 3018:
Derflinger, Gerhard; Hörmann, Wolfgang; Leydold, Josef (2010). "Random variate generation by numerical inversion when only the density is known".
3291: 3286: 1620: 2981: 2029: 2683: 2562:
is a special case of that available for any quantile function whose second derivative exists. In general the equation for a quantile,
1438:, is yet another way of prescribing a probability distribution. It is the reciprocal of the pdf composed with the quantile function. 2949: 2894: 1858:
polynomial when ν is even. In other cases the quantile functions may be developed as power series. The simple cases are as follows:
1495: 1132: 717:, which is equivalent to the previous probability statement in the special case that the distribution is continuous. Note that the 721:
can be replaced by the minimum function, since the distribution function is right-continuous and weakly monotonically increasing.
3356: 1952: 1415: 436: 112: 91: 3443: 2821: 1552: 1475: 3184: 1772: 875: 3350: 1407: 449: 2272: 2190:{\displaystyle q={\frac {\cos \left({\frac {1}{3}}\arccos \left({\sqrt {\alpha }}\,\right)\right)}{\sqrt {\alpha }}}\!} 815:
is continuous and strictly monotonically increasing, then the inequalities can be replaced by equalities, and we have:
2831: 1852: 435:
In the general case of distribution functions that are not strictly monotonic and therefore do not permit an inverse
3432: 2811: 117: 3344: 1511: 1503: 1411: 598:
It is often standard to choose the lowest value, which can equivalently be written as (using right-continuity of
258: 3333: 1839:
approach. From this solutions of arbitrarily high accuracy may be developed (see Steinbrecher and Shaw, 2008).
1446: 941: 67: 3211:
Shaw, W.T. (2006). "Sampling Student's T distribution – Use of the inverse cumulative distribution function".
1875: 177: 1718: 1406:
The quantile function is one way of prescribing a probability distribution, and it is an alternative to the
2853:"Of quantiles and expectiles: Consistent scoring functions, Choquet representations, and forecast rankings" 1339: 335: 1560: 1515: 1491: 1283: 1224: 2206: 1590: 1548: 1507: 1467: 1463: 1084: 769: 3270:"Applying series expansion to the inverse beta distribution to find percentiles of the F-distribution" 733: 2263: 1527: 1499: 718: 1067:{\displaystyle F(x;\lambda )={\begin{cases}1-e^{-\lambda x}&x\geq 0,\\0&x<0.\end{cases}}} 1000: 2559: 2459: 2394: 1867: 1586: 1580: 1556: 1471: 1466:, are focusing increasing attention on methods based on quantile functions, as they work well with 1458: 43: 398: 3404: 3164: 3129: 3073: 2864: 1568: 1400: 2945: 2920: 2358: 1564: 1523: 1487: 821: 725: 3087:
Baumgarten, Christoph; Patel, Tirth (2022). "Automatic random variate generation in Python".
1547:
Quantile functions may also be characterized as solutions of non-linear ordinary and partial
3277: 3247: 3238:
Karvanen, J. (2006). "Estimation of quantile mixtures via L-moments and trimmed L-moments".
3220: 3156: 3121: 3092: 3027: 2993: 2874: 2816: 1519: 1490:, such as the exponential distribution above, which is one of the few distributions where a 1442: 1427: 862: 326: 3147:
Wichura, M.J. (1988). "Algorithm AS241: The Percentage Points of the Normal Distribution".
2497: 2432: 3372: 3188: 2524: 869:
behaves as an "almost sure left inverse" for the distribution function, in the sense that
168: 63: 2901: 3394: 3389: 953: 3301: 17: 3469: 1514:). When the cdf itself has a closed-form expression, one can always use a numerical 3133: 1836: 1457:
Monte-Carlo simulations employ quantile functions to produce non-uniform random or
699:{\displaystyle Q(p)\ =\ \inf \left\{\ x\in \mathbb {R} \ :\ p\leq F(x)\ \right\}~.} 439:, the quantile is a (potentially) set valued functional of a distribution function 3181: 3097: 709:
Here we capture the fact that the quantile function returns the minimum value of
51: 3296: 3269: 3251: 3384: 3182:
An algorithm for computing the inverse normal cumulative distribution function
3125: 3045: 2667:{\displaystyle {\frac {d^{2}Q}{dp^{2}}}=H(Q)\left({\frac {dQ}{dp}}\right)^{2}} 79: 55: 3059: 27:
Statistical function that defines the quantiles of a probability distribution
3454: 3031: 1702:{\displaystyle {\frac {d^{2}w}{dp^{2}}}=w\left({\frac {dw}{dp}}\right)^{2}} 3281: 3224: 2997: 1589:
is perhaps the most important case. Because the normal distribution is a
3449: 3419: 3414: 3409: 3399: 2826: 1526:
series of books. Algorithms for common distributions are built into many
1450: 1212: 71: 3074:"Random Number Generators (Scipy.stats.sampling) — SciPy v1.13.0 Manual" 1835:
This equation may be solved by several methods, including the classical
30: 3200:
Computational Finance: Differential Equations for Monte Carlo Recycling
3168: 2879: 2852: 70:
is less than or equal to an input probability value. Intuitively, the
2099:{\displaystyle Q(p)=\operatorname {sign} (p-1/2)\,2\,{\sqrt {q-1}}\!} 1606:
A non-linear ordinary differential equation for the normal quantile,
1594: 1275: 35: 3160: 2784:{\displaystyle H(x)=-{\frac {f'(x)}{f(x)}}=-{\frac {d}{dx}}\ln f(x)} 2869: 3199: 2982:"Continuous random variate generation by fast numerical inversion" 1541: 1441:
Consider a statistical application where a user needs to know key
397: 29: 3060:"Runuran: R Interface to the 'UNU.RAN' Random Variate Generators" 1399:
Quantile functions are used in both statistical applications and
957: 3305: 414:
values. The quantile function does the opposite: it gives the
1197:{\displaystyle Q(p;\lambda )={\frac {-\ln(1-p)}{\lambda }},\!} 3112:
Steinbrecher, G.; Shaw, W.T. (2008). "Quantile mechanics".
2013:{\displaystyle Q(p)=2(p-1/2){\sqrt {\frac {2}{\alpha }}}\!} 1060: 3046:"UNU.RAN - Universal Non-Uniform RANdom number generators" 2558:
The non-linear ordinary differential equation given for
2554:
Non-linear differential equations for quantile functions
713:
from amongst all those values whose c.d.f value exceeds
1449:
of an observation whose distribution is known; see the
2851:
Ehm, W.; Gneiting, T.; Jordan, A.; Krüger, F. (2016).
1602:
Ordinary differential equation for the normal quantile
1434:. The derivative of the quantile function, namely the 111:
With reference to a continuous and strictly monotonic
2686: 2579: 2527: 2500: 2462: 2435: 2397: 2361: 2275: 2209: 2118: 2032: 1955: 1878: 1825:{\displaystyle w'\left(1/2\right)={\sqrt {2\pi }}.\,} 1775: 1721: 1623: 1342: 1286: 1227: 1135: 1087: 973: 940:
For example, the cumulative distribution function of
878: 824: 772: 736: 611: 452: 338: 261: 180: 120: 3089:
Proceedings of the 21st Python in Science Conference
3020:
ACM Transactions on Modeling and Computer Simulation
2986:
ACM Transactions on Modeling and Computer Simulation
2266:, distributions can be defined as quantile mixtures 1486:
The evaluation of quantile functions often involves
3292:
New Methods for Managing "Student's" T Distribution
924:{\displaystyle Q{\bigl (}\ F(X)\ {\bigr )}=X\quad } 724:The quantile is the unique function satisfying the 2783: 2666: 2542: 2513: 2486: 2448: 2421: 2383: 2343: 2243: 2189: 2098: 2012: 1935: 1824: 1758: 1701: 1383: 1324: 1265: 1196: 1118: 1066: 923: 857:In general, even though the distribution function 846: 800: 758: 698: 588:{\displaystyle Q(p)\ =\ {\boldsymbol {\biggl }}~.} 587: 387: 314: 212: 159: 3268:Abernathy, Roger W. and Smith, Robert P. (1993) * 2677:augmented by suitable boundary conditions, where 2344:{\displaystyle Q(p)=\sum _{i=1}^{m}a_{i}Q_{i}(p)} 2240: 2186: 2095: 2009: 1932: 1193: 1081:) is derived by finding the value of Q for which 574: 476: 633: 532: 484: 284: 402:The cumulative distribution function (shown as 3476:Functions related to probability distributions 3317: 2942:Statistical Modelling with Quantile Functions 909: 884: 8: 3240:Computational Statistics & Data Analysis 1581:Normal distribution § Quantile function 160:{\displaystyle F_{X}\colon \mathbb {R} \to } 315:{\displaystyle F_{X}(x):=\Pr(X\leq x)=p\,,} 3324: 3310: 3302: 2980:Hörmann, Wolfgang; Leydold, Josef (2003). 1571:distributions have been given and solved. 3096: 2878: 2868: 2748: 2705: 2685: 2658: 2634: 2605: 2587: 2580: 2578: 2526: 2505: 2499: 2494:are the model parameters. The parameters 2461: 2440: 2434: 2396: 2366: 2360: 2326: 2316: 2306: 2295: 2274: 2208: 2167: 2160: 2139: 2125: 2117: 2082: 2081: 2077: 2066: 2031: 1997: 1986: 1954: 1918: 1877: 1821: 1808: 1792: 1774: 1755: 1733: 1720: 1693: 1669: 1649: 1631: 1624: 1622: 1380: 1369: 1358: 1341: 1321: 1313: 1302: 1285: 1262: 1254: 1243: 1226: 1157: 1134: 1098: 1086: 1013: 995: 972: 908: 907: 883: 882: 877: 835: 823: 771: 735: 651: 650: 610: 573: 572: 524: 475: 474: 451: 381: 363: 358: 337: 308: 266: 260: 206: 205: 179: 135: 134: 125: 119: 3297:ACM Algorithm 396: Student's t-Quantiles 1936:{\displaystyle Q(p)=\tan(\pi (p-1/2))\!} 1430:of its cumulative distribution function 240:. In terms of the distribution function 213:{\displaystyle Q\colon \to \mathbb {R} } 107:Strictly monotonic distribution function 88:inverse cumulative distribution function 3114:European Journal of Applied Mathematics 2843: 1426:, of a probability distribution is the 2925:: CS1 maint: archived copy as title ( 2918: 1759:{\displaystyle w\left(1/2\right)=0,\,} 1077:The quantile function for Exponential( 1712:with the centre (initial) conditions 1384:{\displaystyle -\ln(1/4)/\lambda .\,} 388:{\displaystyle Q(p)=F_{X}^{-1}(p)\,.} 7: 3155:(3). Blackwell Publishing: 477–484. 1325:{\displaystyle -\ln(1/2)/\lambda \,} 1266:{\displaystyle -\ln(3/4)/\lambda \,} 426:in red is a horizontal line segment. 2244:{\displaystyle \alpha =4p(1-p).\!} 1119:{\displaystyle 1-e^{-\lambda Q}=p} 801:{\displaystyle \quad p\leq F(x)~.} 25: 3287:Refinement of the Normal Quantile 1494:can be found (others include the 1474:or quasi-Monte-Carlo methods and 422:values. Note that the portion of 3371: 3357:cumulative distribution function 3213:Journal of Computational Finance 1862:ν = 1 (Cauchy distribution) 1416:cumulative distribution function 759:{\displaystyle Q(p)\leq x\quad } 525: 113:cumulative distribution function 92:cumulative distribution function 3444:probability-generating function 1553:ordinary differential equations 920: 773: 755: 2967:Monte Carlo methods in finance 2822:Probability integral transform 2778: 2772: 2736: 2730: 2722: 2716: 2696: 2690: 2626: 2620: 2537: 2531: 2378: 2372: 2338: 2332: 2285: 2279: 2234: 2222: 2074: 2054: 2042: 2036: 1994: 1974: 1965: 1959: 1929: 1926: 1906: 1900: 1888: 1882: 1842: 1540:Python subpackage sampling in 1476:Monte Carlo methods in finance 1366: 1352: 1310: 1296: 1251: 1237: 1181: 1169: 1151: 1139: 989: 977: 901: 895: 789: 783: 746: 740: 679: 673: 621: 615: 555: 549: 507: 501: 462: 456: 378: 372: 348: 342: 299: 287: 278: 272: 202: 199: 187: 154: 142: 139: 1: 2487:{\displaystyle i=1,\ldots ,m} 2422:{\displaystyle i=1,\ldots ,m} 1598:have been developed by Shaw. 431:General distribution function 96:inverse distribution function 3351:probability density function 3098:10.25080/majora-212e5952-007 1408:probability density function 418:values as a function of the 410:values as a function of the 2429:are quantile functions and 1470:techniques based on either 228:so that the probability of 3492: 3433:moment-generating function 3252:10.1016/j.csda.2005.09.014 2812:Inverse transform sampling 1865: 1850: 1578: 3428: 3380: 3369: 3345:probability mass function 3340: 3334:probability distributions 3126:10.1017/S0956792508007341 2521:must be selected so that 2264:the mixtures of densities 1436:quantile density function 1422:. The quantile function, 1412:probability mass function 232:being less or equal than 2944:. Taylor & Francis. 2384:{\displaystyle Q_{i}(p)} 1853:Student's t-distribution 1447:statistical significance 1334:third quartile (p = 3/4) 1219:first quartile (p = 1/4) 1211: < 1. The 865:, the quantile function 847:{\displaystyle Q=F^{-1}} 443:, given by the interval 325:which can be written as 244:, the quantile function 174:, the quantile function 3439:characteristic function 3276:, 9 (4), 478–480 3274:ACM Trans. Math. Softw. 3032:10.1145/1842722.1842723 2570:), may be given. It is 1614:), may be given. It is 1420:characteristic function 62:outputs the value of a 3007:– via WU Vienna. 2940:Gilchrist, W. (2000). 2832:Rank–size distribution 2785: 2668: 2544: 2515: 2488: 2450: 2423: 2385: 2345: 2311: 2245: 2191: 2100: 2014: 1937: 1826: 1760: 1703: 1549:differential equations 1516:root-finding algorithm 1492:closed-form expression 1385: 1326: 1267: 1198: 1120: 1068: 925: 861:may fail to possess a 848: 802: 760: 700: 589: 427: 389: 316: 214: 161: 84:percent-point function 47: 18:Percent point function 3282:10.1145/168173.168387 3225:10.21314/JCF.2006.150 2998:10.1145/945511.945517 2786: 2669: 2545: 2516: 2514:{\displaystyle a_{i}} 2489: 2451: 2449:{\displaystyle a_{i}} 2424: 2386: 2346: 2291: 2246: 2192: 2101: 2015: 1938: 1851:Further information: 1827: 1761: 1704: 1591:location-scale family 1555:for the cases of the 1464:computational finance 1386: 1327: 1268: 1199: 1121: 1069: 948:(i.e. intensity 926: 863:left or right inverse 849: 803: 761: 701: 590: 401: 390: 317: 224:to a threshold value 215: 162: 33: 3187:May 5, 2007, at the 2965:Jaeckel, P. (2002). 2684: 2577: 2543:{\displaystyle Q(p)} 2525: 2498: 2460: 2433: 2395: 2359: 2273: 2207: 2116: 2030: 1953: 1876: 1773: 1719: 1621: 1528:statistical software 1506:(which includes the 1459:pseudorandom numbers 1340: 1284: 1225: 1133: 1085: 971: 876: 822: 770: 734: 609: 450: 336: 259: 178: 118: 2560:normal distribution 1868:Cauchy distribution 1587:normal distribution 1575:Normal distribution 1401:Monte Carlo methods 726:Galois inequalities 371: 76:percentile function 44:normal distribution 3405:standard deviation 3149:Applied Statistics 3091:. pp. 46–51. 3062:. 17 January 2023. 2880:10.1111/rssb.12154 2857:J. R. Stat. Soc. B 2781: 2664: 2540: 2511: 2484: 2446: 2419: 2381: 2341: 2241: 2187: 2096: 2010: 1933: 1822: 1756: 1699: 1537:R library Runuran 1534:C library UNU.RAN 1381: 1322: 1263: 1207:for 0 ≤  1194: 1116: 1064: 1059: 921: 844: 798: 756: 696: 585: 428: 385: 354: 312: 248:returns the value 210: 157: 48: 3463: 3462: 3363:quantile function 2907:on March 24, 2012 2761: 2740: 2652: 2612: 2258:Quantile mixtures 2184: 2183: 2165: 2147: 2093: 2007: 2006: 1816: 1687: 1656: 1524:Numerical Recipes 1488:numerical methods 1443:percentage points 1188: 906: 891: 794: 692: 684: 663: 657: 643: 632: 626: 581: 571: 531: 523: 483: 473: 467: 60:quantile function 40:quantile function 16:(Redirected from 3483: 3375: 3326: 3319: 3312: 3303: 3256: 3255: 3235: 3229: 3228: 3208: 3202: 3197: 3191: 3179: 3173: 3172: 3144: 3138: 3137: 3109: 3103: 3102: 3100: 3084: 3078: 3077: 3070: 3064: 3063: 3056: 3050: 3049: 3042: 3036: 3035: 3015: 3009: 3008: 3006: 3004: 2977: 2971: 2970: 2962: 2956: 2955: 2937: 2931: 2930: 2924: 2916: 2914: 2912: 2906: 2900:. Archived from 2899: 2891: 2885: 2884: 2882: 2872: 2848: 2817:Percentage point 2790: 2788: 2787: 2782: 2762: 2760: 2749: 2741: 2739: 2725: 2715: 2706: 2673: 2671: 2670: 2665: 2663: 2662: 2657: 2653: 2651: 2643: 2635: 2613: 2611: 2610: 2609: 2596: 2592: 2591: 2581: 2549: 2547: 2546: 2541: 2520: 2518: 2517: 2512: 2510: 2509: 2493: 2491: 2490: 2485: 2455: 2453: 2452: 2447: 2445: 2444: 2428: 2426: 2425: 2420: 2390: 2388: 2387: 2382: 2371: 2370: 2350: 2348: 2347: 2342: 2331: 2330: 2321: 2320: 2310: 2305: 2250: 2248: 2247: 2242: 2196: 2194: 2193: 2188: 2185: 2179: 2178: 2177: 2173: 2172: 2168: 2166: 2161: 2148: 2140: 2126: 2105: 2103: 2102: 2097: 2094: 2083: 2070: 2019: 2017: 2016: 2011: 2008: 1999: 1998: 1990: 1942: 1940: 1939: 1934: 1922: 1831: 1829: 1828: 1823: 1817: 1809: 1804: 1800: 1796: 1783: 1765: 1763: 1762: 1757: 1745: 1741: 1737: 1708: 1706: 1705: 1700: 1698: 1697: 1692: 1688: 1686: 1678: 1670: 1657: 1655: 1654: 1653: 1640: 1636: 1635: 1625: 1520:bisection method 1390: 1388: 1387: 1382: 1373: 1362: 1331: 1329: 1328: 1323: 1317: 1306: 1272: 1270: 1269: 1264: 1258: 1247: 1203: 1201: 1200: 1195: 1189: 1184: 1158: 1125: 1123: 1122: 1117: 1109: 1108: 1073: 1071: 1070: 1065: 1063: 1062: 1024: 1023: 930: 928: 927: 922: 913: 912: 904: 889: 888: 887: 868: 860: 853: 851: 850: 845: 843: 842: 814: 811:If the function 807: 805: 804: 799: 792: 766:if and only if 765: 763: 762: 757: 719:infimum function 716: 712: 705: 703: 702: 697: 690: 689: 685: 682: 661: 655: 654: 641: 630: 624: 601: 594: 592: 591: 586: 579: 578: 577: 569: 568: 564: 529: 528: 521: 520: 516: 481: 480: 479: 471: 465: 442: 394: 392: 391: 386: 370: 362: 321: 319: 318: 313: 271: 270: 219: 217: 216: 211: 209: 166: 164: 163: 158: 138: 130: 129: 21: 3491: 3490: 3486: 3485: 3484: 3482: 3481: 3480: 3466: 3465: 3464: 3459: 3424: 3376: 3367: 3336: 3330: 3265: 3263:Further reading 3260: 3259: 3237: 3236: 3232: 3210: 3209: 3205: 3198: 3194: 3189:Wayback Machine 3180: 3176: 3161:10.2307/2347330 3146: 3145: 3141: 3111: 3110: 3106: 3086: 3085: 3081: 3072: 3071: 3067: 3058: 3057: 3053: 3044: 3043: 3039: 3017: 3016: 3012: 3002: 3000: 2979: 2978: 2974: 2964: 2963: 2959: 2952: 2939: 2938: 2934: 2917: 2910: 2908: 2904: 2897: 2895:"Archived copy" 2893: 2892: 2888: 2850: 2849: 2845: 2840: 2808: 2753: 2726: 2708: 2707: 2682: 2681: 2644: 2636: 2630: 2629: 2601: 2597: 2583: 2582: 2575: 2574: 2556: 2523: 2522: 2501: 2496: 2495: 2458: 2457: 2436: 2431: 2430: 2393: 2392: 2362: 2357: 2356: 2322: 2312: 2271: 2270: 2262:Analogously to 2260: 2205: 2204: 2159: 2155: 2138: 2134: 2127: 2114: 2113: 2028: 2027: 1951: 1950: 1874: 1873: 1870: 1855: 1849: 1788: 1784: 1776: 1771: 1770: 1729: 1725: 1717: 1716: 1679: 1671: 1665: 1664: 1645: 1641: 1627: 1626: 1619: 1618: 1604: 1583: 1577: 1484: 1397: 1338: 1337: 1282: 1281: 1223: 1222: 1215:are therefore: 1159: 1131: 1130: 1094: 1083: 1082: 1058: 1057: 1046: 1040: 1039: 1025: 1009: 996: 969: 968: 938: 874: 873: 866: 858: 831: 820: 819: 812: 768: 767: 732: 731: 714: 710: 640: 636: 607: 606: 599: 539: 535: 491: 487: 448: 447: 440: 433: 334: 333: 262: 257: 256: 220:maps its input 176: 175: 169:random variable 121: 116: 115: 109: 104: 64:random variable 28: 23: 22: 15: 12: 11: 5: 3489: 3487: 3479: 3478: 3468: 3467: 3461: 3460: 3458: 3457: 3452: 3447: 3441: 3436: 3429: 3426: 3425: 3423: 3422: 3417: 3412: 3407: 3402: 3397: 3392: 3390:central moment 3387: 3381: 3378: 3377: 3370: 3368: 3366: 3365: 3360: 3354: 3348: 3341: 3338: 3337: 3331: 3329: 3328: 3321: 3314: 3306: 3300: 3299: 3294: 3289: 3284: 3264: 3261: 3258: 3257: 3246:(2): 947–956. 3230: 3203: 3192: 3174: 3139: 3104: 3079: 3065: 3051: 3037: 3034:. Art. No. 18. 3010: 2992:(4): 347–362. 2972: 2957: 2950: 2932: 2886: 2863:(3): 505–562. 2842: 2841: 2839: 2836: 2835: 2834: 2829: 2824: 2819: 2814: 2807: 2804: 2792: 2791: 2780: 2777: 2774: 2771: 2768: 2765: 2759: 2756: 2752: 2747: 2744: 2738: 2735: 2732: 2729: 2724: 2721: 2718: 2714: 2711: 2704: 2701: 2698: 2695: 2692: 2689: 2675: 2674: 2661: 2656: 2650: 2647: 2642: 2639: 2633: 2628: 2625: 2622: 2619: 2616: 2608: 2604: 2600: 2595: 2590: 2586: 2555: 2552: 2539: 2536: 2533: 2530: 2508: 2504: 2483: 2480: 2477: 2474: 2471: 2468: 2465: 2443: 2439: 2418: 2415: 2412: 2409: 2406: 2403: 2400: 2380: 2377: 2374: 2369: 2365: 2353: 2352: 2340: 2337: 2334: 2329: 2325: 2319: 2315: 2309: 2304: 2301: 2298: 2294: 2290: 2287: 2284: 2281: 2278: 2259: 2256: 2252: 2251: 2239: 2236: 2233: 2230: 2227: 2224: 2221: 2218: 2215: 2212: 2198: 2197: 2182: 2176: 2171: 2164: 2158: 2154: 2151: 2146: 2143: 2137: 2133: 2130: 2124: 2121: 2107: 2106: 2092: 2089: 2086: 2080: 2076: 2073: 2069: 2065: 2062: 2059: 2056: 2053: 2050: 2047: 2044: 2041: 2038: 2035: 2025: 2021: 2020: 2005: 2002: 1996: 1993: 1989: 1985: 1982: 1979: 1976: 1973: 1970: 1967: 1964: 1961: 1958: 1948: 1944: 1943: 1931: 1928: 1925: 1921: 1917: 1914: 1911: 1908: 1905: 1902: 1899: 1896: 1893: 1890: 1887: 1884: 1881: 1866:Main article: 1864: 1863: 1848: 1841: 1833: 1832: 1820: 1815: 1812: 1807: 1803: 1799: 1795: 1791: 1787: 1782: 1779: 1767: 1766: 1754: 1751: 1748: 1744: 1740: 1736: 1732: 1728: 1724: 1710: 1709: 1696: 1691: 1685: 1682: 1677: 1674: 1668: 1663: 1660: 1652: 1648: 1644: 1639: 1634: 1630: 1603: 1600: 1579:Main article: 1576: 1573: 1545: 1544: 1538: 1535: 1483: 1480: 1418:(cdf) and the 1396: 1393: 1392: 1391: 1379: 1376: 1372: 1368: 1365: 1361: 1357: 1354: 1351: 1348: 1345: 1335: 1332: 1320: 1316: 1312: 1309: 1305: 1301: 1298: 1295: 1292: 1289: 1279: 1273: 1261: 1257: 1253: 1250: 1246: 1242: 1239: 1236: 1233: 1230: 1220: 1205: 1204: 1192: 1187: 1183: 1180: 1177: 1174: 1171: 1168: 1165: 1162: 1156: 1153: 1150: 1147: 1144: 1141: 1138: 1115: 1112: 1107: 1104: 1101: 1097: 1093: 1090: 1075: 1074: 1061: 1056: 1053: 1050: 1047: 1045: 1042: 1041: 1038: 1035: 1032: 1029: 1026: 1022: 1019: 1016: 1012: 1008: 1005: 1002: 1001: 999: 994: 991: 988: 985: 982: 979: 976: 954:expected value 937: 936:Simple example 934: 933: 932: 931:almost surely. 919: 916: 911: 903: 900: 897: 894: 886: 881: 855: 854: 841: 838: 834: 830: 827: 809: 808: 797: 791: 788: 785: 782: 779: 776: 754: 751: 748: 745: 742: 739: 707: 706: 695: 688: 681: 678: 675: 672: 669: 666: 660: 653: 649: 646: 639: 635: 629: 623: 620: 617: 614: 596: 595: 584: 576: 567: 563: 560: 557: 554: 551: 548: 545: 542: 538: 534: 527: 519: 515: 512: 509: 506: 503: 500: 497: 494: 490: 486: 478: 470: 464: 461: 458: 455: 432: 429: 396: 395: 384: 380: 377: 374: 369: 366: 361: 357: 353: 350: 347: 344: 341: 329:of the c.d.f. 323: 322: 311: 307: 304: 301: 298: 295: 292: 289: 286: 283: 280: 277: 274: 269: 265: 208: 204: 201: 198: 195: 192: 189: 186: 183: 156: 153: 150: 147: 144: 141: 137: 133: 128: 124: 108: 105: 103: 100: 94:or c.d.f.) or 66:such that its 26: 24: 14: 13: 10: 9: 6: 4: 3: 2: 3488: 3477: 3474: 3473: 3471: 3456: 3453: 3451: 3448: 3445: 3442: 3440: 3437: 3434: 3431: 3430: 3427: 3421: 3418: 3416: 3413: 3411: 3408: 3406: 3403: 3401: 3398: 3396: 3393: 3391: 3388: 3386: 3383: 3382: 3379: 3374: 3364: 3361: 3358: 3355: 3352: 3349: 3346: 3343: 3342: 3339: 3335: 3327: 3322: 3320: 3315: 3313: 3308: 3307: 3304: 3298: 3295: 3293: 3290: 3288: 3285: 3283: 3279: 3275: 3271: 3267: 3266: 3262: 3253: 3249: 3245: 3241: 3234: 3231: 3226: 3222: 3218: 3214: 3207: 3204: 3201: 3196: 3193: 3190: 3186: 3183: 3178: 3175: 3170: 3166: 3162: 3158: 3154: 3150: 3143: 3140: 3135: 3131: 3127: 3123: 3120:(2): 87–112. 3119: 3115: 3108: 3105: 3099: 3094: 3090: 3083: 3080: 3075: 3069: 3066: 3061: 3055: 3052: 3047: 3041: 3038: 3033: 3029: 3025: 3021: 3014: 3011: 2999: 2995: 2991: 2987: 2983: 2976: 2973: 2968: 2961: 2958: 2953: 2951:1-58488-174-7 2947: 2943: 2936: 2933: 2928: 2922: 2903: 2896: 2890: 2887: 2881: 2876: 2871: 2866: 2862: 2858: 2854: 2847: 2844: 2837: 2833: 2830: 2828: 2825: 2823: 2820: 2818: 2815: 2813: 2810: 2809: 2805: 2803: 2801: 2797: 2775: 2769: 2766: 2763: 2757: 2754: 2750: 2745: 2742: 2733: 2727: 2719: 2712: 2709: 2702: 2699: 2693: 2687: 2680: 2679: 2678: 2659: 2654: 2648: 2645: 2640: 2637: 2631: 2623: 2617: 2614: 2606: 2602: 2598: 2593: 2588: 2584: 2573: 2572: 2571: 2569: 2565: 2561: 2553: 2551: 2534: 2528: 2506: 2502: 2481: 2478: 2475: 2472: 2469: 2466: 2463: 2441: 2437: 2416: 2413: 2410: 2407: 2404: 2401: 2398: 2375: 2367: 2363: 2335: 2327: 2323: 2317: 2313: 2307: 2302: 2299: 2296: 2292: 2288: 2282: 2276: 2269: 2268: 2267: 2265: 2257: 2255: 2237: 2231: 2228: 2225: 2219: 2216: 2213: 2210: 2203: 2202: 2201: 2180: 2174: 2169: 2162: 2156: 2152: 2149: 2144: 2141: 2135: 2131: 2128: 2122: 2119: 2112: 2111: 2110: 2090: 2087: 2084: 2078: 2071: 2067: 2063: 2060: 2057: 2051: 2048: 2045: 2039: 2033: 2026: 2023: 2022: 2003: 2000: 1991: 1987: 1983: 1980: 1977: 1971: 1968: 1962: 1956: 1949: 1946: 1945: 1923: 1919: 1915: 1912: 1909: 1903: 1897: 1894: 1891: 1885: 1879: 1872: 1871: 1869: 1861: 1860: 1859: 1854: 1847:-distribution 1846: 1840: 1838: 1818: 1813: 1810: 1805: 1801: 1797: 1793: 1789: 1785: 1780: 1777: 1769: 1768: 1752: 1749: 1746: 1742: 1738: 1734: 1730: 1726: 1722: 1715: 1714: 1713: 1694: 1689: 1683: 1680: 1675: 1672: 1666: 1661: 1658: 1650: 1646: 1642: 1637: 1632: 1628: 1617: 1616: 1615: 1613: 1609: 1601: 1599: 1596: 1592: 1588: 1582: 1574: 1572: 1570: 1566: 1562: 1558: 1554: 1550: 1543: 1539: 1536: 1533: 1532: 1531: 1529: 1525: 1521: 1517: 1513: 1509: 1505: 1501: 1497: 1493: 1489: 1481: 1479: 1477: 1473: 1469: 1465: 1460: 1455: 1452: 1448: 1444: 1439: 1437: 1433: 1429: 1425: 1421: 1417: 1413: 1409: 1404: 1402: 1394: 1377: 1374: 1370: 1363: 1359: 1355: 1349: 1346: 1343: 1336: 1333: 1318: 1314: 1307: 1303: 1299: 1293: 1290: 1287: 1280: 1277: 1274: 1259: 1255: 1248: 1244: 1240: 1234: 1231: 1228: 1221: 1218: 1217: 1216: 1214: 1210: 1190: 1185: 1178: 1175: 1172: 1166: 1163: 1160: 1154: 1148: 1145: 1142: 1136: 1129: 1128: 1127: 1113: 1110: 1105: 1102: 1099: 1095: 1091: 1088: 1080: 1054: 1051: 1048: 1043: 1036: 1033: 1030: 1027: 1020: 1017: 1014: 1010: 1006: 1003: 997: 992: 986: 983: 980: 974: 967: 966: 965: 963: 959: 955: 951: 947: 945: 935: 917: 914: 898: 892: 879: 872: 871: 870: 864: 839: 836: 832: 828: 825: 818: 817: 816: 795: 786: 780: 777: 774: 752: 749: 743: 737: 730: 729: 728: 727: 722: 720: 693: 686: 676: 670: 667: 664: 658: 647: 644: 637: 627: 618: 612: 605: 604: 603: 582: 565: 561: 558: 552: 546: 543: 540: 536: 517: 513: 510: 504: 498: 495: 492: 488: 468: 459: 453: 446: 445: 444: 438: 430: 425: 421: 417: 413: 409: 405: 400: 382: 375: 367: 364: 359: 355: 351: 345: 339: 332: 331: 330: 328: 309: 305: 302: 296: 293: 290: 281: 275: 267: 263: 255: 254: 253: 251: 247: 243: 239: 235: 231: 227: 223: 196: 193: 190: 184: 181: 173: 170: 151: 148: 145: 131: 126: 122: 114: 106: 101: 99: 97: 93: 89: 85: 81: 77: 73: 69: 65: 61: 57: 53: 45: 41: 37: 32: 19: 3362: 3273: 3243: 3239: 3233: 3219:(4): 37–73. 3216: 3212: 3206: 3195: 3177: 3152: 3148: 3142: 3117: 3113: 3107: 3088: 3082: 3068: 3054: 3040: 3023: 3019: 3013: 3001:. Retrieved 2989: 2985: 2975: 2966: 2960: 2941: 2935: 2909:. Retrieved 2902:the original 2889: 2860: 2856: 2846: 2799: 2795: 2793: 2676: 2567: 2563: 2557: 2354: 2261: 2253: 2199: 2108: 1856: 1844: 1837:power series 1834: 1711: 1611: 1607: 1605: 1584: 1546: 1518:such as the 1512:log-logistic 1504:Tukey lambda 1485: 1468:multivariate 1456: 1440: 1435: 1431: 1423: 1405: 1398: 1395:Applications 1208: 1206: 1078: 1076: 961: 949: 943: 942:Exponential( 939: 856: 810: 723: 708: 597: 434: 423: 419: 415: 411: 407: 406:) gives the 403: 324: 249: 245: 241: 237: 233: 229: 225: 221: 171: 110: 95: 87: 83: 75: 59: 49: 39: 3026:(4): 1–25. 1542:scipy.stats 1482:Calculation 90:(after the 78:(after the 68:probability 52:probability 3385:raw moment 3332:Theory of 2870:1503.08195 2838:References 2024:ν = 4 1947:ν = 2 1843:Student's 1510:) and the 252:such that 102:Definition 80:percentile 56:statistics 3455:combinant 2911:March 25, 2767:⁡ 2746:− 2703:− 2476:… 2411:… 2293:∑ 2229:− 2211:α 2181:α 2163:α 2153:⁡ 2132:⁡ 2088:− 2061:− 2052:⁡ 2004:α 1981:− 1913:− 1904:π 1898:⁡ 1814:π 1410:(pdf) or 1375:λ 1350:⁡ 1344:− 1319:λ 1294:⁡ 1288:− 1278:(p = 2/4) 1260:λ 1235:⁡ 1229:− 1213:quartiles 1186:λ 1176:− 1167:⁡ 1161:− 1149:λ 1103:λ 1100:− 1092:− 1031:≥ 1018:λ 1015:− 1007:− 987:λ 960:) 1/ 837:− 778:≤ 750:≤ 668:≤ 648:∈ 559:≤ 544:: 496:: 365:− 294:≤ 203:→ 185:: 140:→ 132:: 3470:Category 3450:cumulant 3420:L-moment 3415:kurtosis 3410:skewness 3400:variance 3185:Archived 2921:cite web 2827:Quantile 2806:See also 2713:′ 1781:′ 1508:logistic 1451:quantile 72:quantile 3169:2347330 3134:6899308 3003:17 June 1561:Student 1500:Weibull 1496:uniform 1428:inverse 327:inverse 42:of the 38:is the 3167:  3132:  2948:  2355:where 2150:arccos 2109:where 1595:probit 1557:normal 1551:. The 1502:, the 1498:, the 1472:copula 1414:, the 1276:median 905:  890:  793:  691:  683:  662:  656:  642:  631:  625:  580:  570:  530:  522:  482:  472:  466:  437:c.d.f. 58:, the 36:probit 3446:(pgf) 3435:(mgf) 3359:(cdf) 3353:(pdf) 3347:(pmf) 3165:JSTOR 3130:S2CID 2905:(PDF) 2898:(PDF) 2865:arXiv 1569:gamma 964:) is 167:of a 3395:mean 3005:2024 2946:ISBN 2927:link 2913:2012 2794:and 2200:and 2049:sign 1585:The 1567:and 1565:beta 1052:< 958:mean 952:and 511:< 424:F(x) 404:F(x) 54:and 34:The 3278:doi 3248:doi 3221:doi 3157:doi 3122:doi 3093:doi 3028:doi 2994:doi 2875:doi 2129:cos 1895:tan 634:inf 533:sup 485:sup 236:is 82:), 50:In 3472:: 3272:, 3244:51 3242:. 3215:. 3163:. 3153:37 3151:. 3128:. 3118:19 3116:. 3024:20 3022:. 2990:13 2988:. 2984:. 2923:}} 2919:{{ 2873:. 2861:78 2859:. 2855:. 2764:ln 2456:, 2391:, 1563:, 1559:, 1478:. 1403:. 1347:ln 1291:ln 1232:ln 1164:ln 1126:: 1055:0. 602:) 285:Pr 282::= 98:. 86:, 3325:e 3318:t 3311:v 3280:: 3254:. 3250:: 3227:. 3223:: 3217:9 3171:. 3159:: 3136:. 3124:: 3101:. 3095:: 3076:. 3048:. 3030:: 2996:: 2969:. 2954:. 2929:) 2915:. 2883:. 2877:: 2867:: 2800:x 2798:( 2796:ƒ 2779:) 2776:x 2773:( 2770:f 2758:x 2755:d 2751:d 2743:= 2737:) 2734:x 2731:( 2728:f 2723:) 2720:x 2717:( 2710:f 2700:= 2697:) 2694:x 2691:( 2688:H 2660:2 2655:) 2649:p 2646:d 2641:Q 2638:d 2632:( 2627:) 2624:Q 2621:( 2618:H 2615:= 2607:2 2603:p 2599:d 2594:Q 2589:2 2585:d 2568:p 2566:( 2564:Q 2538:) 2535:p 2532:( 2529:Q 2507:i 2503:a 2482:m 2479:, 2473:, 2470:1 2467:= 2464:i 2442:i 2438:a 2417:m 2414:, 2408:, 2405:1 2402:= 2399:i 2379:) 2376:p 2373:( 2368:i 2364:Q 2351:, 2339:) 2336:p 2333:( 2328:i 2324:Q 2318:i 2314:a 2308:m 2303:1 2300:= 2297:i 2289:= 2286:) 2283:p 2280:( 2277:Q 2238:. 2235:) 2232:p 2226:1 2223:( 2220:p 2217:4 2214:= 2175:) 2170:) 2157:( 2145:3 2142:1 2136:( 2123:= 2120:q 2091:1 2085:q 2079:2 2075:) 2072:2 2068:/ 2064:1 2058:p 2055:( 2046:= 2043:) 2040:p 2037:( 2034:Q 2001:2 1995:) 1992:2 1988:/ 1984:1 1978:p 1975:( 1972:2 1969:= 1966:) 1963:p 1960:( 1957:Q 1930:) 1927:) 1924:2 1920:/ 1916:1 1910:p 1907:( 1901:( 1892:= 1889:) 1886:p 1883:( 1880:Q 1845:t 1819:. 1811:2 1806:= 1802:) 1798:2 1794:/ 1790:1 1786:( 1778:w 1753:, 1750:0 1747:= 1743:) 1739:2 1735:/ 1731:1 1727:( 1723:w 1695:2 1690:) 1684:p 1681:d 1676:w 1673:d 1667:( 1662:w 1659:= 1651:2 1647:p 1643:d 1638:w 1633:2 1629:d 1612:p 1610:( 1608:w 1432:F 1424:Q 1378:. 1371:/ 1367:) 1364:4 1360:/ 1356:1 1353:( 1315:/ 1311:) 1308:2 1304:/ 1300:1 1297:( 1256:/ 1252:) 1249:4 1245:/ 1241:3 1238:( 1209:p 1191:, 1182:) 1179:p 1173:1 1170:( 1155:= 1152:) 1146:; 1143:p 1140:( 1137:Q 1114:p 1111:= 1106:Q 1096:e 1089:1 1079:λ 1049:x 1044:0 1037:, 1034:0 1028:x 1021:x 1011:e 1004:1 998:{ 993:= 990:) 984:; 981:x 978:( 975:F 962:λ 956:( 950:λ 946:) 944:λ 918:X 915:= 910:) 902:) 899:X 896:( 893:F 885:( 880:Q 867:Q 859:F 840:1 833:F 829:= 826:Q 813:F 796:. 790:) 787:x 784:( 781:F 775:p 753:x 747:) 744:p 741:( 738:Q 715:p 711:x 694:. 687:} 680:) 677:x 674:( 671:F 665:p 659:: 652:R 645:x 638:{ 628:= 622:) 619:p 616:( 613:Q 600:F 583:. 575:] 566:} 562:p 556:) 553:x 550:( 547:F 541:x 537:{ 526:, 518:} 514:p 508:) 505:x 502:( 499:F 493:x 489:{ 477:[ 469:= 463:) 460:p 457:( 454:Q 441:F 420:p 416:q 412:q 408:p 383:. 379:) 376:p 373:( 368:1 360:X 356:F 352:= 349:) 346:p 343:( 340:Q 310:, 306:p 303:= 300:) 297:x 291:X 288:( 279:) 276:x 273:( 268:X 264:F 250:x 246:Q 242:F 238:p 234:x 230:X 226:x 222:p 207:R 200:] 197:1 194:, 191:0 188:[ 182:Q 172:X 155:] 152:1 149:, 146:0 143:[ 136:R 127:X 123:F 46:. 20:)

Index

Percent point function

probit
normal distribution
probability
statistics
random variable
probability
quantile
percentile
cumulative distribution function
cumulative distribution function
random variable
inverse

c.d.f.
infimum function
Galois inequalities
left or right inverse
Exponential(λ)
expected value
mean
quartiles
median
Monte Carlo methods
probability density function
probability mass function
cumulative distribution function
characteristic function
inverse

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