Scribd Logo
Upload

Welcome to Scribd!

  • 69 views

    Prueba de Anderson Darling

    Uploaded by

    sandysaryono
    The document describes the Anderson Darling test, which is used to test if a sample comes from a specific distribution. It is a modification of the Kolmogorov-Smirnov test that gives more weight to the tails of the distribution. The steps of the test are provided: 1) obtain sample data, 2) order data ascending, 3) order data descending, 4) establish the null hypothesis, 5) calculate test statistics, 6) compare test statistic to critical values. An example is shown applying the test to determine if a data set comes from a normal distribution. The test statistic is calculated to be less than the critical value, so the null hypothesis that the data comes from a normal distribution is accepted.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Prueba de Anderson Darling

    Uploaded by

    sandysaryono
    69 views8 pages
    The document describes the Anderson Darling test, which is used to test if a sample comes from a specific distribution. It is a modification of the Kolmogorov-Smirnov test that gives more weight to the tails of the distribution. The steps of the test are provided: 1) obtain sample data, 2) order data ascending, 3) order data descending, 4) establish the null hypothesis, 5) calculate test statistics, 6) compare test statistic to critical values. An example is shown applying the test to determine if a data set comes from a normal distribution. The test statistic is calculated to be less than the critical value, so the null hypothesis that the data comes from a normal distribution is accepted.

    Original Description:

    Un TEMA con ejemplo de la prueba de anderson darling

    Original Title

    PRUEBA DE ANDERSON DARLING

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes the Anderson Darling test, which is used to test if a sample comes from a specific distribution. It is a modification of the Kolmogorov-Smirnov test that gives more weight to the tails of the distribution. The steps of the test are provided: 1) obtain sample data, 2) order data ascending, 3) order data descending, 4) establish the null hypothesis, 5) calculate test statistics, 6) compare test statistic to critical values. An example is shown applying the test to determine if a data set comes from a normal distribution. The test statistic is calculated to be less than the critical value, so the null hypothesis that the data comes from a normal distribution is accepted.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    69 views8 pages

    Prueba de Anderson Darling

    Uploaded by

    sandysaryono
    The document describes the Anderson Darling test, which is used to test if a sample comes from a specific distribution. It is a modification of the Kolmogorov-Smirnov test that gives more weight to the tails of the distribution. The steps of the test are provided: 1) obtain sample data, 2) order data ascending, 3) order data descending, 4) establish the null hypothesis, 5) calculate test statistics, 6) compare test statistic to critical values. An example is shown applying the test to determine if a data set comes from a normal distribution. The test statistic is calculated to be less than the critical value, so the null hypothesis that the data comes from a normal distribution is accepted.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 8

    PRUEBA DE

    ANDERSON DARLING
    Jacqueline Ramos C.
    PASOS

    Es usada para probar si una


    muestra viene de una 1. Obtener n datos
    distribución específica. Es una 2. Organizar los datos en forma
    modificación de la prueba de ascendente Yi
    Kolmogorov Smirnov donde se 3. Organizar los datos en forma
    le da mas peso a las colas de descendente Yn+1-i
    la distribución. 4. Establecer la hipótesis nula
    5. Calcular PEA(Yi) , PEA(Yn+1-i)
    6. Calcular el valor del estadístico A

    2
    Si el estadístico es menor al valor de tablas, entonces se acepta
    la hipótesis nula

    3
    EJEMPLO. Determinar la distribución a la que se ajusta el siguiente conjunto de
    datos:

    12.99 19.67 22.26 10.20 13.29


    46.76 13.72 14.75 8.71 9.77
    14.70 51.90 25.82 10.31 10.27
    24.46 14.14 15.11 13.13 11.38
    13.08 10.00 42.99 14.56 12.24
    36.07 20.62 13.17 23.08 41.04

    4
    valor max 14.751
    valor min 5.633
    m 7
    amplitud 1.302571

    Intervalo FO
    5.633 6.935571 6
    6.935571 8.238143 5
    8.238143 9.540714 17
    9.540714 10.84329 7
    10.84329 12.14586 5
    12.14586 13.44843 5
    13.44843 14.751 5
    50

    Ho: Distribución normal

    media 9.78602
    desviacion 2.326741
    i Yi Yn+1-i 2i-1 z de yi PEA(Yi) z de Yn+1-i 1-PEA(Yn+1-i) ln(PEA(Yi)) LN(1-PEA(Yn+1-i))
    1 5.633 14.751 1 -1.7849 0.0371 2.1339 0.0164 -3.2931 -4.1089 -7.4020
    2 5.764 14.389 3 -1.7286 0.0419 1.9783 0.0239 -3.1715 -3.7319 -20.7102
    3 6.306 13.944 5 -1.4957 0.0674 1.7870 0.0370 -2.6975 -3.2978 -29.9766
    4 6.515 13.838 7 -1.4058 0.0799 1.7415 0.0408 -2.5272 -3.1991 -40.0837
    5 6.619 13.466 9 -1.3611 0.0867 1.5816 0.0569 -2.4449 -2.8670 -47.8070
    6 6.775 13.323 11 -1.2941 0.0978 1.5201 0.0642 -2.3247 -2.7452 -55.7681
    7 6.957 12.685 13 -1.2159 0.1120 1.2459 0.1064 -2.1891 -2.2406 -57.5864
    8 7.112 12.553 15 -1.1493 0.1252 1.1892 0.1172 -2.0776 -2.1441 -63.3254
    9 7.445 12.397 17 -1.0061 0.1572 1.1222 0.1309 -1.8504 -2.0333 -66.0236
    10 7.827 12.19 19 -0.8420 0.1999 1.0332 0.1508 -1.6099 -1.8921 -66.5380
    …
    …

    40 11.804 8.045 79 0.8673 0.8071 -0.7483 0.7728 -0.2143 -0.2577 -37.2852


    41 12.19 7.827 81 1.0332 0.8492 -0.8420 0.8001 -0.1634 -0.2230 -31.3012
    42 12.397 7.445 83 1.1222 0.8691 -1.0061 0.8428 -0.1403 -0.1710 -25.8370
    43 12.553 7.112 85 1.1892 0.8828 -1.1493 0.8748 -0.1246 -0.1338 -21.9658
    44 12.685 6.957 87 1.2459 0.8936 -1.2159 0.8880 -0.1125 -0.1188 -20.1224
    45 13.323 6.775 89 1.5201 0.9358 -1.2941 0.9022 -0.0664 -0.1029 -15.0705
    46 13.466 6.619 91 1.5816 0.9431 -1.3611 0.9133 -0.0586 -0.0907 -13.5845
    47 13.838 6.515 93 1.7415 0.9592 -1.4058 0.9201 -0.0417 -0.0833 -11.6169
    48 13.944 6.306 95 1.7870 0.9630 -1.4957 0.9326 -0.0377 -0.0697 -10.2043
    49 14.389 5.764 97 1.9783 0.9761 -1.7286 0.9581 -0.0242 -0.0428 -6.5071
    50 14.751 5.633 99 2.1339 0.9836 -1.7849 0.9629 -0.0166 -0.0378 -5.3864
    -2530.214889

    estadistico A 0.604297782
    Valor de tablas:

    Comparando

    como 0.604297782 es menor al valor de tablas,


    entonces SE ACEPTA la hipótesis de que los números
    provienen de una distribución normal

    7
    PRACTICA

    Ajustar los siguientes datos a alguna distribución de probabilidad usar alfa 10%

    13.29 12.99 19.67 22.26 10.20


    9.77 46.76 13.72 14.75 8.71
    10.27 14.70 51.90 25.82 10.31
    11.38 24.46 14.14 15.11 13.13
    12.24 13.08 10.00 42.99 14.56
    41.04 36.07 20.62 13.17 23.08

    You might also like