[R-br] dúvida - modelo

Quinta Agosto 16 22:47:34 BRT 2012

Oi

Obrigado por sempre ajudar Benilton.

Deixa eu tentar explicar o objetivo do trabalho.

Na verdade que vai trabalhar com esses dados é um amigo no doutorado.

A variável resposta é Atividade Física e ele quer modelar, na verdade, seria gasto energético por AF, os 0 são indivíduos que não praticam AF, ou seja, tem gasto 0, acima desse valor, são gastos energéticos de atividades físicas quaisquer...
Os dados não são normais e quando tento normalizar com boxcox não roda por conta dos 0.
Quando fiz um histograma dessa variável, me pareceu que os mesmos apresentam distribuição geométrica, parei por aí...

Obrigado mais uma vez.

Enviado via iPod

Em 16/08/2012, às 22:11, Benilton Carvalho <beniltoncarvalho em gmail.com> escreveu:

> modelagem envolve muito mais que os dados com os quais vc esta'
> trabalhando... na verdade, o processo gerador dos dados e' q e'
> relevante... vc precisa entender a "historia" dos dados...
> 
> sobre os seus dados, pergunte-se: 1) eles sao sempre inteiros? 2) se
> sim, eles representam o numero de tentativas (de alguma coisa) ate' o
> primeiro sucesso (ou falha)? 3) se sim, cada uma dessas tentativas tem
> a mesma chance de sucesso (ou falha)?
> 
> se sua resposta for "nao" para qq uma das perguntas acima, e' pouco
> provavel que uma geometrica seja a resposta q vc procura...
> 
> b
> 
> 2012/8/17 Sérgio Henrique almeida da silva ju <sergio.edfisica em gmail.com>:
>> Olá
>> 
>> Estou modelando um conjunto de dados e a variável resposta parece ter
>> distribuição geométrica, qual é a melhor forma de modelar esses dados?
>> 
>> Abraços
>> 
>> 
>> dput(dados)
>> structure(list(AFtotal = structure(c(103L, 76L, 58L, 105L, 125L,
>> 163L, 106L, 146L, 227L, 187L, 5L, 117L, 147L, 39L, 232L, 2L,
>> 67L, 150L, 46L, 118L, 18L, 171L, 157L, 226L, 80L, 96L, 142L,
>> 162L, 128L, 152L, 109L, 10L, 108L, 102L, 47L, 208L, 122L, 78L,
>> 176L, 3L, 196L, 40L, 209L, 119L, 179L, 29L, 1L, 81L, 190L, 24L,
>> 204L, 28L, 92L, 136L, 104L, 85L, 91L, 15L, 141L, 13L, 11L, 121L,
>> 112L, 210L, 31L, 6L, 70L, 156L, 114L, 68L, 107L, 112L, 211L,
>> 167L, 1L, 215L, 88L, 213L, 194L, 136L, 220L, 199L, 136L, 9L,
>> 136L, 118L, 38L, 192L, 42L, 62L, 155L, 19L, 79L, 48L, 138L, 172L,
>> 134L, 231L, 231L, 216L, 136L, 195L, 130L, 101L, 217L, 205L, 174L,
>> 36L, 111L, 140L, 106L, 99L, 22L, 218L, 170L, 7L, 8L, 136L, 19L,
>> 62L, 25L, 123L, 1L, 137L, 220L, 97L, 52L, 169L, 190L, 216L, 1L,
>> 53L, 172L, 100L, 79L, 86L, 201L, 61L, 198L, 17L, 218L, 67L, 106L,
>> 193L, 1L, 138L, 164L, 200L, 190L, 133L, 129L, 74L, 186L, 72L,
>> 41L, 177L, 48L, 113L, 50L, 229L, 178L, 106L, 124L, 75L, 14L,
>> 148L, 218L, 191L, 19L, 219L, 30L, 1L, 71L, 112L, 172L, 21L, 221L,
>> 16L, 203L, 27L, 69L, 143L, 82L, 207L, 223L, 32L, 54L, 60L, 59L,
>> 1L, 97L, 149L, 63L, 132L, 212L, 214L, 1L, 1L, 165L, 77L, 20L,
>> 73L, 26L, 168L, 160L, 233L, 139L, 1L, 188L, 95L, 175L, 206L,
>> 93L, 144L, 135L, 56L, 6L, 189L, 197L, 45L, 30L, 35L, 1L, 4L,
>> 136L, 154L, 89L, 65L, 184L, 225L, 230L, 151L, 1L, 173L, 43L,
>> 44L, 33L, 182L, 34L, 153L, 1L, 1L, 90L, 106L, 110L, 98L, 1L,
>> 219L, 1L, 1L, 116L, 49L, 64L, 57L, 131L, 185L, 145L, 17L, 94L,
>> 1L, 180L, 1L, 120L, 159L, 51L, 172L, 181L, 183L, 7L, 87L, 106L,
>> 228L, 55L, 126L, 157L, 218L, 158L, 115L, 133L, 222L, 84L, 166L,
>> 1L, 37L, 1L, 1L, 202L, 66L, 12L, 1L, 83L, 23L, 110L, 233L, 231L,
>> 161L, 106L, 224L, 127L), .Label = c("0,00", "1.031,00", "1.035,00",
>> "1.042,50", "1.064,00", "1.080,00", "1.120,00", "1.155,00", "1.171,50",
>> "1.177,50", "1.179,00", "1.200,00", "1.204,50", "1.210,00", "1.218,00",
>> "1.220,00", "1.230,00", "1.233,00", "1.260,00", "1.275,00", "1.287,00",
>> "1.290,00", "1.314,00", "1.320,00", "1.356,00", "1.359,50", "1.400,00",
>> "1.410,00", "1.418,00", "1.440,00", "1.466,00", "1.480,00", "1.500,00",
>> "1.566,00", "1.572,00", "1.590,00", "1.611,00", "1.652,00", "1.680,00",
>> "1.722,00", "1.734,00", "1.740,00", "1.752,00", "1.800,00", "1.818,00",
>> "1.821,00", "1.833,00", "1.836,00", "1.854,00", "1.872,00", "1.900,00",
>> "1.920,00", "1.938,00", "10.080,00", "10.230,00", "10.557,00",
>> "10.680,00", "11.238,00", "12.000,00", "12.270,00", "12.420,00",
>> "120,00", "13.644,00", "13.860,00", "14.590,00", "150,00", "165,00",
>> "180,00", "2.002,50", "2.060,00", "2.078,00", "2.100,00", "2.120,00",
>> "2.157,00", "2.320,00", "2.332,50", "2.376,00", "2.382,00", "2.430,00",
>> "2.457,00", "2.475,00", "2.504,00", "2.520,00", "2.526,00", "2.532,00",
>> "2.556,00", "2.572,00", "2.580,00", "2.616,00", "2.640,00", "2.645,00",
>> "2.655,00", "2.706,00", "2.739,00", "2.745,00", "2.767,50", "2.784,00",
>> "2.798,00", "2.820,00", "2.841,00", "2.895,00", "2.937,00", "2.940,00",
>> "2.965,00", "21.515,00", "240,00", "247,50", "259,00", "26.862,00",
>> "264,00", "273,00", "297,00", "3.019,00", "3.035,00", "3.036,00",
>> "3.282,00", "3.306,00", "3.360,00", "3.497,00", "3.504,00", "3.510,00",
>> "3.540,00", "3.600,00", "3.615,00", "3.660,00", "3.710,00", "3.840,00",
>> "3.892,50", "3.900,00", "3.960,00", "3.977,00", "3.996,00", "300,00",
>> "318,00", "330,00", "360,00", "372,00", "396,00", "4.020,00",
>> "4.086,00", "4.092,00", "4.100,00", "4.140,00", "4.158,00", "4.200,00",
>> "4.219,50", "4.331,00", "4.359,00", "4.554,00", "4.618,00", "4.680,00",
>> "4.770,00", "407,50", "410,00", "420,00", "424,00", "480,00",
>> "495,00", "5.052,00", "5.160,00", "5.340,00", "5.622,00", "5.688,00",
>> "5.715,00", "5.745,00", "5.832,00", "5.940,00", "5.958,00", "510,00",
>> "516,00", "537,00", "540,00", "565,50", "570,00", "575,00", "576,00",
>> "585,00", "6.097,50", "6.285,00", "6.495,00", "6.556,00", "6.576,00",
>> "6.720,00", "6.746,00", "6.828,00", "639,00", "645,00", "655,00",
>> "66,00", "660,00", "678,00", "690,00", "693,00", "696,00", "7.056,00",
>> "7.074,00", "7.200,00", "7.236,00", "7.245,00", "7.500,00", "7.600,00",
>> "7.680,00", "7.860,00", "7.920,00", "720,00", "738,00", "754,00",
>> "754,50", "780,00", "795,00", "8.268,00", "8.467,50", "8.592,00",
>> "8.755,00", "8.820,00", "80,00", "801,00", "810,00", "812,00",
>> "840,00", "876,00", "882,00", "895,50", "9.120,00", "9.150,00",
>> "9.310,00", "9.508,00", "924,00", "930,00", "933,00", "960,00",
>> "99,00", "990,00"), class = "factor")), .Names = "AFtotal", class =
>> "data.frame", row.names = c(NA,
>> -299L))
>> 
>> 
>> --
>> Sérgio Henrique Almeida da Silva Junior
>> Doutorando em Epidemiologia em Saúde Pública
>> Escola Nacional de Saúde Pública Sérgio Arouca - ENSP/FIOCRUZ
>> http://lattes.cnpq.br/1611345552843383
>> Tel: (21) 94429486/78101651 id: 123*20942
>> 
>> 
>> 
>> _______________________________________________
>> R-br mailing list
>> R-br em listas.c3sl.ufpr.br
>> https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br
>> Leia o guia de postagem (http://www.leg.ufpr.br/r-br-guia) e forneça código
>> mínimo reproduzível.
> _______________________________________________
> R-br mailing list
> R-br em listas.c3sl.ufpr.br
> https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br
> Leia o guia de postagem (http://www.leg.ufpr.br/r-br-guia) e forneça código mínimo reproduzível.