<div dir="ltr"><div>Walmes fiz algumas pequenas alterações no código que me enviou para conseguir obter uma saída para o modelo lme. Desculpe ficar perguntando, mas não conheco bem essa técnica. <br><br></div><div>A questão é: Baseada na saída abaixo, como eu a torno interpretável? Faço a média desses valores? Estou pensado em como avaliar se o modelo ajustou. É possivel com essa metodologia obter estimativas de MSE, MAPE, CCC, etc. Estou confuso sobre isso.<br><br></div><div>abraços<br></div><div><br><br></div>#Dados para o CMR abaixo:<br><br><div><div>modelo3<-lme(fixed= CMS~PM+GPD+FDN+I(FDN^2),data=da,random=~1|Estudo,na.action=na.omit,method="REML")<br><br>L <- vector(mode="list", length=nrow(da))<br>for(i in 1:nrow(da)){<br> L[[i]] <-lme(fixed= CMS~PM+GPD+FDN+I(FDN^2),random=~1|Estudo,na.action=na.omit,method="REML", data=da[-i,])<br>}<br><br>## Medidas de deviance leave-one-out.<br><br>sapply(L, FUN=function(x){-2*logLik(x)}) #Deviance<br><br>## log-veross leave-one-out.<br>sapply(L, logLik)<br><br>da<-structure(list(Estudo = structure(c(1L, 1L, 1L, 1L, 1L, 2L, 2L, <br>2L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 6L, 6L, 6L, <br>6L, 6L, 6L, 6L, 6L, 7L, 7L, 7L, 7L, 8L, 8L, 8L, 8L, 9L, 9L, 9L, <br>10L, 10L, 10L, 10L, 11L, 11L, 11L, 11L, 12L, 12L, 12L, 12L, 13L, <br>13L, 13L, 13L, 13L, 13L, 14L, 14L, 14L), .Label = c("1", "2", <br>"3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", <br>"15", "16", "17", "18", "19", "20", "21", "22", "23", "24", "25", <br>"26", "27", "28", "29", "30", "31", "32", "33", "34", "35", "36", <br>"37", "38", "39", "40", "41", "42", "43", "44"), class = "factor"), <br> NANIMAL = c(20L, 20L, 20L, 20L, 20L, 18L, 18L, 18L, 20L, <br> 20L, 20L, 20L, 20L, 20L, 20L, 20L, 20L, 18L, 18L, 18L, 16L, <br> 16L, 16L, 16L, 16L, 16L, 16L, 16L, 24L, 24L, 24L, 24L, 20L, <br> 20L, 20L, 20L, 18L, 18L, 18L, 24L, 24L, 24L, 24L, 24L, 24L, <br> 24L, 24L, 20L, 20L, 20L, 20L, 18L, 18L, 18L, 18L, 18L, 18L, <br> 10L, 10L, 10L), GENOTIPO = structure(c(2L, 2L, 2L, 2L, 2L, <br> 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, <br> 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, <br> 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, <br> 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = c("1", <br> "2"), class = "factor"), SEXO = structure(c(1L, 1L, 1L, 1L, <br> 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, <br> 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, <br> 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, <br> 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = c("1", <br> "2"), class = "factor"), VOL = structure(c(1L, 1L, 1L, 1L, <br> 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, <br> 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, <br> 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, <br> 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L), .Label = c("1", <br> "2"), class = "factor"), PVI = c(25, 25, 25, 25, 25, 20.17, <br> 20.1, 20.47, 21.3, 22, 22.9, 22.9, 18.25, 18.25, 22.47, 19.05, <br> 16.63, 22.62, 22.8, 22.95, 26.5, 26.5, 26.5, 26.5, 26.5, <br> 26.5, 26.5, 26.5, 19.1, 19.16, 19.96, 19.96, 20, 20, 20, <br> 20, 17, 16.25, 15.58, 16.4, 16.1, 16.6, 16.5, 18.74, 18.05, <br> 18.82, 18.15, 35.47, 35.7, 35.53, 35.66, 22.6, 22.6, 22.6, <br> 22.6, 22.6, 22.6, 15, 25, 35), PVF = c(36.8, 35.6, 33.88, <br> 35.8, 32.48, 31.87, 31.1, 31.9, 27.1, 27.6, 28.7, 31, 30.46, <br> 30.46, 32.15, 24.28, 19.62, 37.67, 39.48, 39.43, 30.15, 31.04, <br> 31.88, 31.92, 30.11, 29.44, 29.48, 30.87, 33.5, 32.66, 30.41, <br> 32.16, 25.23, 26.9, 28.65, 29.23, 24.66, 25.08, 27.75, 28.7, <br> 29.6, 31.1, 30.9, 31.7, 32.42, 34.97, 32.65, 44.21, 46.18, <br> 44.62, 45.55, 30, 27.73, 29.06, 32.7, 30.1, 28.8, 25, 35, <br> 45), PV = c(30.9, 30.3, 29.44, 30.4, 28.74, 26.02, 25.6, <br> 26.19, 24.2, 24.8, 25.8, 26.95, 24.36, 24.36, 27.31, 21.67, <br> 18.13, 30.15, 31.14, 31.19, 28.33, 28.77, 29.19, 29.21, 28.31, <br> 27.97, 27.99, 28.69, 26.3, 25.91, 25.19, 26.06, 22.62, 23.45, <br> 24.33, 24.62, 20.83, 20.67, 21.67, 22.55, 22.85, 23.85, 23.7, <br> 25.22, 25.24, 26.9, 25.4, 39.84, 40.94, 40.08, 40.61, 26.3, <br> 25.17, 25.83, 27.65, 26.35, 25.7, 20, 30, 40), PM = c(13.11, <br> 12.91, 12.64, 12.95, 12.41, 11.52, 11.38, 11.58, 10.91, 11.11, <br> 11.45, 11.83, 10.96, 10.96, 11.95, 10.04, 8.78, 12.87, 13.18, <br> 13.2, 12.28, 12.42, 12.56, 12.56, 12.27, 12.16, 12.17, 12.39, <br> 11.61, 11.48, 11.24, 11.53, 10.37, 10.66, 10.95, 11.05, 9.75, <br> 9.69, 10.04, 10.35, 10.45, 10.79, 10.74, 11.25, 11.26, 11.81, <br> 11.31, 15.86, 16.18, 15.93, 16.09, 11.61, 11.24, 11.46, 12.06, <br> 11.63, 11.41, 9.46, 12.82, 15.91), GPD = c(295, 265, 222, <br> 270, 187, 123, 137, 191, 86.5, 100.4, 92.1, 140.5, 195.83, <br> 195.83, 161.25, 87.08, 49.75, 268.75, 297.85, 294.28, 87, <br> 108, 128, 129, 86, 70, 71, 104, 206, 186, 149, 174, 93.4, <br> 123.2, 154.4, 164.8, 88.12, 101.53, 139.84, 195, 214.5, 229.9, <br> 228, 245, 271, 305, 274, 202, 243, 211, 212, 117.5, 81.5, <br> 102.6, 160.3, 119, 98.4, 205, 261, 183), D.EXP = c(40, 40, <br> 40, 40, 40, 98, 97, 77, 77, 77, 77, 77, 60, 60, 60, 60, 60, <br> 56, 56, 56, 42, 42, 42, 42, 42, 42, 42, 42, 70, 70, 70, 70, <br> 56, 56, 56, 56, 87, 87, 87, 63, 63, 63, 63, 53, 53, 53, 53, <br> 40, 40, 40, 40, 63, 63, 63, 63, 63, 63, 52, 55, 65), CMS = c(1490, <br> 1540, 1390, 1580, 1440, 900, 870, 870, 422, 501, 525, 533, <br> 371, 1377, 975.7, 596.7, 448.9, 1260, 1220, 1170, 889.56, <br> 842.96, 945.76, 1051.56, 888.93, 819.52, 906.88, 1044.11, <br> 946.8, 1236, 1126, 1192, 648.6, 764.4, 960, 941, 845.28, <br> 845.61, 780.59, 820.8, 830.6, 870, 887.8, 1098, 1159, 1289, <br> 1173, 1200, 1340, 1300, 1230, 1001, 987, 861, 896, 1037, <br> 785, 707.6, 977.3, 1032.9), FDN = c(41.52, 42.99, 53.49, <br> 51.2, 56.45, 58.3, 44.23, 29.96, 59.18, 55.36, 51.55, 47.73, <br> 43.91, 52.3, 52.8, 47.7, 44.9, 34.5, 34.5, 34.5, 64.21, 60.77, <br> 61.62, 59.32, 64.21, 60.77, 61.62, 59.32, 36.4, 41.6, 49.6, <br> 49.6, 54.7, 55.9, 54, 56.6, 35.46, 34.87, 34.26, 44.2, 43.44, <br> 42.4, 41.7, 43.36, 39.62, 35.88, 32.15, 31.46, 35.55, 39.54, <br> 44.07, 12.37, 25.47, 39.16, 12.37, 25.47, 39.16, 21.41, 21.41, <br> 21.41), CMSPV = c(48.22, 50.83, 47.21, 51.97, 50.1, 34.59, <br> 33.98, 33.23, 31.52, 35.31, 37.15, 37.12, 15.23, 56.54, 35.73, <br> 27.54, 24.77, 41.8, 39.18, 37.51, 31.4, 29.3, 32.4, 36, 31.4, <br> 29.3, 32.4, 36.39, 36, 47.7, 44.7, 45.74, 28.67, 32.6, 39.46, <br> 38.22, 40.58, 40.91, 36.02, 36.4, 36.35, 36.48, 37.46, 43.54, <br> 45.92, 47.92, 46.18, 30.12, 32.73, 32.44, 30.29, 38.06, 39.21, <br> 33.33, 32.41, 39.35, 30.54, 35.38, 32.58, 25.82), CMSPM = c(113.69, <br> 119.24, 109.98, 122.04, 116.01, 78.12, 76.44, 75.16, 64.76, <br> 72.67, 78.02, 78.42, 60.33, 125.6, 81.67, 59.42, 51.1, 97.94, <br> 92.55, 88.65, 72.44, 67.87, 75.3, 83.72, 72.45, 67.39, 74.52, <br> 84.27, 105.57, 107.63, 100.16, 103.35, 62.54, 71.73, 87.65, <br> 85.15, 87.68, 88.32, 79.27, 79.32, 79.47, 80.61, 82.65, 97.56, <br> 102.94, 109.14, 103.67, 80.17, 88.22, 86.12, 81.77, 78.46, <br> 83.78, 75.14, 74.31, 89.17, 68.77, 74.82, 76.24, 64.94)), .Names = c("Estudo", <br>"NANIMAL", "GENOTIPO", "SEXO", "VOL", "PVI", "PVF", "PV", "PM", <br>"GPD", "D.EXP", "CMS", "FDN", "CMSPV", "CMSPM"), row.names = c(NA, <br>60L), class = "data.frame")<br><br></div></div></div><div class="gmail_extra"><br><div class="gmail_quote">Em 1 de junho de 2015 23:18, Fernando Antonio de souza <span dir="ltr"><<a href="mailto:nandodesouza@gmail.com" target="_blank">nandodesouza@gmail.com</a>></span> escreveu:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div><div>Obrigado, Walmes,<br><br></div>Irei testar aqui e retornarei a lista. Irei estudar mais sobre o tema. Hoje li alguns tutoriais sobre o assunto e realmente tenho a impressão de ser muito simples. Mas não tinha ideía o quão simples era para implementá-la.<br><br></div>Abçs<br></div><div class="gmail_extra"><br><div class="gmail_quote"><div><div class="h5">Em 1 de junho de 2015 21:23, walmes . <span dir="ltr"><<a href="mailto:walmeszeviani@gmail.com" target="_blank">walmeszeviani@gmail.com</a>></span> escreveu:<br></div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5"><div dir="ltr"><div class="gmail_default" style="font-family:trebuchet ms,sans-serif">Eu posso estar enganado então me corrijam em caso afirmativo. O leave-ONE-out é mais simples do que imagina. Grosseiramente falando, crie um laço for para deixar um caso de fora e ajuste o modelo. Depois estude uma medida de ajuste. No caso de lm, não requer de fato fazer a exaustiva tarefa de ajustar n modelos (deixando um caso de fora), pois se chega as medidas leave-one-out por projeções matriciais, etc. Mas num caso mais geral é algo como:<br><br></div><div class="gmail_default"><span style="font-family:monospace,monospace">da ## seu data.frame<br></span></div><div class="gmail_default"><span style="font-family:monospace,monospace">L <- vector(mode="list", length=nrow(da))<br></span></div><div class="gmail_default"><span style="font-family:monospace,monospace">for(i in 1:nrow(da)){<br></span></div><div class="gmail_default"><span style="font-family:monospace,monospace"> L[[i]] <- sua_funcao_R(..., data=da[-i,])<br></span></div><div class="gmail_default"><span style="font-family:monospace,monospace">}<br><br></span></div><div class="gmail_default"><span style="font-family:monospace,monospace">## Medidas de deviance leave-one-out.<br></span></div><div class="gmail_default"><span style="font-family:monospace,monospace">sapply(L, deviance)<br><br></span><div class="gmail_default"><span style="font-family:monospace,monospace">## log-veross leave-one-out.<br></span></div><span style="font-family:monospace,monospace">sapply(L, logLik)<br><br></span></div><div class="gmail_default" style="font-family:trebuchet ms,sans-serif"><span style="font-family:monospace,monospace">Código não testado.</span><br><br>À disposição.<span><font color="#888888"><br></font></span></div><span><font color="#888888"><div class="gmail_default" style="font-family:trebuchet ms,sans-serif">Walmes.<br></div><div class="gmail_default" style="font-family:trebuchet ms,sans-serif"><br></div></font></span></div>
<br></div></div><span class="">_______________________________________________<br>
R-br mailing list<br>
<a href="mailto:R-br@listas.c3sl.ufpr.br" target="_blank">R-br@listas.c3sl.ufpr.br</a><br>
<a href="https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br" target="_blank">https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br</a><br>
Leia o guia de postagem (<a href="http://www.leg.ufpr.br/r-br-guia" target="_blank">http://www.leg.ufpr.br/r-br-guia</a>) e forneça código mínimo reproduzível.<br></span></blockquote></div><span class=""><br><br clear="all"><br>-- <br><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr">=======================================================================<br>Fernando Souza<br>Zootecnista, DSc. Produção Animal<br><a href="mailto:e-mail%3Anandodesouza@gmail.com" target="_blank">e-mail:nandodesouza@gmail.com</a><br><a href="https://producaoanimalcomr.wordpress.com/" target="_blank">https://producaoanimalcomr.wordpress.com/</a><br>========================================================================<br></div></div></div></div></div></div></div></div>
</span></div>
</blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr">=======================================================================<br>Fernando Souza<br>Zootecnista, DSc. Produção Animal<br><a href="mailto:e-mail%3Anandodesouza@gmail.com" target="_blank">e-mail:nandodesouza@gmail.com</a><br><a href="https://producaoanimalcomr.wordpress.com/" target="_blank">https://producaoanimalcomr.wordpress.com/</a><br>========================================================================<br></div></div></div></div></div></div></div></div>
</div>