[R-br] Erro em fit não linear
Michelle Bau Graczyk
mbgraczyk em gmail.com
Quinta Outubro 8 17:12:44 BRT 2015
Caros,boa noite,
estou tentando fazer um fit em um conjunto de dados mas ele me retorna o
erro:
Erro em nls(y ~ func(x, c, a, b), data = dados2, start = guess, trace =
TRUE) :
fator de passos 0.000488281 reduzido abaixo de 'minFactor' de 0.000976562
Alguém, por favor, saberia me dizer como posso concertar isso?
O programa segue abaixo:
> require(lattice)
> l<-1
>
file<-read.table(paste0("/Users/bau/ProjetoU-Shape/AA.N_Novo_MomentosEstatísticosSemestre",l,".txt"),
header=TRUE)
> media<-file[3:389,2]
> media
[1] 24117.500 13168.333 13861.667 18031.667 19581.667 18797.500 15147.500
16699.167 17911.667
[10] 16430.000 15874.167 15470.000 16360.833 18240.000 15018.333 18155.833
14793.333 17348.333
[19] 19142.500 21510.833 17417.500 17450.833 19335.833 14439.167 16995.000
14624.167 18152.500
[28] 18240.000 16124.167 18489.167 14679.167 17181.667 18115.833 15302.500
13445.833 15159.167
[37] 20161.667 15321.667 14252.500 15850.833 13745.000 13915.833 15810.833
13280.833 14680.833
[46] 12119.167 12100.833 16263.333 13210.833 15323.333 12921.667 14374.167
12805.833 13955.000
[55] 12920.833 12016.667 13795.000 14751.667 12735.833 12267.500 12896.667
11947.500 12723.333
[64] 14574.167 12818.333 14187.500 11534.167 12134.167 10910.000 12596.667
11503.333 12515.833
[73] 12709.167 14239.167 11495.000 13163.333 11990.000 10506.667 11984.167
10225.000 11447.500
[82] 11966.667 12115.833 11273.333 10478.333 10375.000 12240.833 11178.333
10699.167 11474.167
[91] 10075.000 12563.333 14194.167 13675.833 12290.000 11357.500 10956.667
12120.000 11663.333
[100] 11078.333 11371.667 10549.167 12766.667 14523.333 10948.333 10809.167
10772.500 11190.000
[109] 9748.333 14768.333 11070.000 10362.500 10942.500 9190.833 12433.333
10302.500 10838.333
[118] 12184.167 8899.167 10073.333 10393.333 9760.000 10390.833 11309.167
9319.167 8183.333
[127] 9531.667 9843.333 8924.167 10070.833 8730.000 10109.167 10507.500
11991.667 8540.833
[136] 11396.667 9405.000 9284.167 9146.667 9640.000 8991.667 8352.500
7969.167 10366.667
[145] 9319.167 8555.000 8578.333 7897.500 8881.667 8842.500 9579.167
9022.500 9914.167
[154] 9933.333 10987.500 8835.000 8473.333 9029.167 8619.167 8782.500
9901.667 8566.667
[163] 8490.833 7475.833 7510.833 8990.000 6865.833 6564.167 7418.333
7356.667 7485.000
[172] 8482.500 9325.000 7989.167 7370.833 7921.667 8243.333 8005.833
8551.667 9119.167
[181] 8760.000 7193.333 9693.333 7742.500 11112.500 7497.500 6685.000
8441.667 7150.000
[190] 6189.167 6760.833 6255.833 8953.333 8413.333 7731.667 6546.667
7179.167 8060.000
[199] 7023.333 9560.000 7071.667 7770.000 6640.000 5910.833 8896.667
7938.333 7098.333
[208] 7282.500 7291.667 8557.500 7326.667 8196.667 12226.667 8380.000
6513.333 9100.833
[217] 6303.333 7525.000 7297.500 7375.833 7224.167 7029.167 9325.833
5923.333 7674.167
[226] 5882.500 8318.333 7263.333 7574.167 7268.333 7530.000 9355.833
8654.167 7426.667
[235] 7794.167 7008.333 7974.167 10410.000 7181.667 7440.000 6345.000
5503.333 6674.167
[244] 9780.000 7740.000 6981.667 5926.667 9352.500 10264.167 8406.667
6532.500 6478.333
[253] 8561.667 8072.500 7700.000 7189.167 6443.333 8120.000 7190.833
6520.833 7539.167
[262] 8013.333 7191.667 7660.000 7276.667 8233.333 7500.833 8582.500
8019.167 7231.667
[271] 10179.167 8445.000 9302.500 7680.833 9114.167 8432.500 7475.000
7779.167 8895.000
[280] 9171.667 9753.333 7490.000 10211.667 11380.000 8524.167 8077.500
10155.000 9406.667
[289] 11199.167 8286.667 9850.000 10032.500 9740.833 7321.667 8494.167
10023.333 9450.000
[298] 10995.833 10166.667 10881.667 8832.500 10015.000 11526.667 11015.833
10857.500 9627.500
[307] 11716.667 10213.333 9765.000 8673.333 8560.833 10153.333 12676.667
11633.333 10933.333
[316] 10492.500 10300.000 9853.333 10602.500 9778.333 9030.000 12785.833
10950.000 11523.333
[325] 12445.000 10896.667 10875.833 11619.167 13154.167 11693.333 12561.667
11002.500 11017.500
[334] 10700.833 14557.500 12236.667 10875.833 11762.500 12705.000 13452.500
11402.500 11232.500
[343] 11739.167 13043.333 11583.333 11468.333 11772.500 13278.333 14787.500
13462.500 13837.500
[352] 13491.667 12455.000 14412.500 14211.667 15135.000 13923.333 15175.000
16048.333 16182.500
[361] 17939.167 17191.667 18522.500 18923.333 17106.667 15255.833 14752.500
15868.333 21636.667
[370] 17807.500 19451.667 17064.167 20163.333 16959.167 17077.500 19470.833
17424.167 18996.667
[379] 24262.500 21454.167 21115.833 18455.000 22215.000 27894.167 28049.167
30302.500 37730.000
> y<-media
> tempo<-c(-193:193/193)
> tempo
[1] -1.000000000 -0.994818653 -0.989637306 -0.984455959 -0.979274611
-0.974093264 -0.968911917
[8] -0.963730570 -0.958549223 -0.953367876 -0.948186528 -0.943005181
-0.937823834 -0.932642487
[15] -0.927461140 -0.922279793 -0.917098446 -0.911917098 -0.906735751
-0.901554404 -0.896373057
[22] -0.891191710 -0.886010363 -0.880829016 -0.875647668 -0.870466321
-0.865284974 -0.860103627
[29] -0.854922280 -0.849740933 -0.844559585 -0.839378238 -0.834196891
-0.829015544 -0.823834197
[36] -0.818652850 -0.813471503 -0.808290155 -0.803108808 -0.797927461
-0.792746114 -0.787564767
[43] -0.782383420 -0.777202073 -0.772020725 -0.766839378 -0.761658031
-0.756476684 -0.751295337
[50] -0.746113990 -0.740932642 -0.735751295 -0.730569948 -0.725388601
-0.720207254 -0.715025907
[57] -0.709844560 -0.704663212 -0.699481865 -0.694300518 -0.689119171
-0.683937824 -0.678756477
[64] -0.673575130 -0.668393782 -0.663212435 -0.658031088 -0.652849741
-0.647668394 -0.642487047
[71] -0.637305699 -0.632124352 -0.626943005 -0.621761658 -0.616580311
-0.611398964 -0.606217617
[78] -0.601036269 -0.595854922 -0.590673575 -0.585492228 -0.580310881
-0.575129534 -0.569948187
[85] -0.564766839 -0.559585492 -0.554404145 -0.549222798 -0.544041451
-0.538860104 -0.533678756
[92] -0.528497409 -0.523316062 -0.518134715 -0.512953368 -0.507772021
-0.502590674 -0.497409326
[99] -0.492227979 -0.487046632 -0.481865285 -0.476683938 -0.471502591
-0.466321244 -0.461139896
[106] -0.455958549 -0.450777202 -0.445595855 -0.440414508 -0.435233161
-0.430051813 -0.424870466
[113] -0.419689119 -0.414507772 -0.409326425 -0.404145078 -0.398963731
-0.393782383 -0.388601036
[120] -0.383419689 -0.378238342 -0.373056995 -0.367875648 -0.362694301
-0.357512953 -0.352331606
[127] -0.347150259 -0.341968912 -0.336787565 -0.331606218 -0.326424870
-0.321243523 -0.316062176
[134] -0.310880829 -0.305699482 -0.300518135 -0.295336788 -0.290155440
-0.284974093 -0.279792746
[141] -0.274611399 -0.269430052 -0.264248705 -0.259067358 -0.253886010
-0.248704663 -0.243523316
[148] -0.238341969 -0.233160622 -0.227979275 -0.222797927 -0.217616580
-0.212435233 -0.207253886
[155] -0.202072539 -0.196891192 -0.191709845 -0.186528497 -0.181347150
-0.176165803 -0.170984456
[162] -0.165803109 -0.160621762 -0.155440415 -0.150259067 -0.145077720
-0.139896373 -0.134715026
[169] -0.129533679 -0.124352332 -0.119170984 -0.113989637 -0.108808290
-0.103626943 -0.098445596
[176] -0.093264249 -0.088082902 -0.082901554 -0.077720207 -0.072538860
-0.067357513 -0.062176166
[183] -0.056994819 -0.051813472 -0.046632124 -0.041450777 -0.036269430
-0.031088083 -0.025906736
[190] -0.020725389 -0.015544041 -0.010362694 -0.005181347 0.000000000
0.005181347 0.010362694
[197] 0.015544041 0.020725389 0.025906736 0.031088083 0.036269430
0.041450777 0.046632124
[204] 0.051813472 0.056994819 0.062176166 0.067357513 0.072538860
0.077720207 0.082901554
[211] 0.088082902 0.093264249 0.098445596 0.103626943 0.108808290
0.113989637 0.119170984
[218] 0.124352332 0.129533679 0.134715026 0.139896373 0.145077720
0.150259067 0.155440415
[225] 0.160621762 0.165803109 0.170984456 0.176165803 0.181347150
0.186528497 0.191709845
[232] 0.196891192 0.202072539 0.207253886 0.212435233 0.217616580
0.222797927 0.227979275
[239] 0.233160622 0.238341969 0.243523316 0.248704663 0.253886010
0.259067358 0.264248705
[246] 0.269430052 0.274611399 0.279792746 0.284974093 0.290155440
0.295336788 0.300518135
[253] 0.305699482 0.310880829 0.316062176 0.321243523 0.326424870
0.331606218 0.336787565
[260] 0.341968912 0.347150259 0.352331606 0.357512953 0.362694301
0.367875648 0.373056995
[267] 0.378238342 0.383419689 0.388601036 0.393782383 0.398963731
0.404145078 0.409326425
[274] 0.414507772 0.419689119 0.424870466 0.430051813 0.435233161
0.440414508 0.445595855
[281] 0.450777202 0.455958549 0.461139896 0.466321244 0.471502591
0.476683938 0.481865285
[288] 0.487046632 0.492227979 0.497409326 0.502590674 0.507772021
0.512953368 0.518134715
[295] 0.523316062 0.528497409 0.533678756 0.538860104 0.544041451
0.549222798 0.554404145
[302] 0.559585492 0.564766839 0.569948187 0.575129534 0.580310881
0.585492228 0.590673575
[309] 0.595854922 0.601036269 0.606217617 0.611398964 0.616580311
0.621761658 0.626943005
[316] 0.632124352 0.637305699 0.642487047 0.647668394 0.652849741
0.658031088 0.663212435
[323] 0.668393782 0.673575130 0.678756477 0.683937824 0.689119171
0.694300518 0.699481865
[330] 0.704663212 0.709844560 0.715025907 0.720207254 0.725388601
0.730569948 0.735751295
[337] 0.740932642 0.746113990 0.751295337 0.756476684 0.761658031
0.766839378 0.772020725
[344] 0.777202073 0.782383420 0.787564767 0.792746114 0.797927461
0.803108808 0.808290155
[351] 0.813471503 0.818652850 0.823834197 0.829015544 0.834196891
0.839378238 0.844559585
[358] 0.849740933 0.854922280 0.860103627 0.865284974 0.870466321
0.875647668 0.880829016
[365] 0.886010363 0.891191710 0.896373057 0.901554404 0.906735751
0.911917098 0.917098446
[372] 0.922279793 0.927461140 0.932642487 0.937823834 0.943005181
0.948186528 0.953367876
[379] 0.958549223 0.963730570 0.968911917 0.974093264 0.979274611
0.984455959 0.989637306
[386] 0.994818653 1.000000000
> x<-tempo
>
> dados <- data.frame(x=x,y=y)
> dados2 <- dados[-c(1,387),]
>
> xyplot(y~x, data=dados, type=c("p","smooth"))
>
> ## Valores iniciais.
>
> guess <- list(c=1.0,a=1.0, b=0.001)
>
> ## Modelo não linear.
> func <- function(x, c,a,b){
+ y <- (atanh((sqrt((x+(b*(x^2)))^2)/c)^2) + a)
+ return(y)
+ }
> with(guess,
+ curve(func(x, c,a,b),min(dados$x), max(dados$x)))
Mensagens de aviso perdidas:
In atanh((sqrt((x + (b * (x^2)))^2)/c)^2) : NaNs produzidos
> abline(v=194, lty=2)
>
>
> ## Sobrepondo dados e função a partir dos valores iniciais.
> plot(y~x, data=dados, xlab="tempo", ylab="volume/média")
>
>
> f <- expression(atanh((sqrt((x+(b*(x^2)))^2)/c)^2) + a)
>
> fit <- nls(y~func(x, c,a,b), data=dados2, start=guess, trace=TRUE)
55389072390 : 1.000 1.000 0.001
6020970037 : -224.49929 10765.82603 49.04659
Erro em nls(y ~ func(x, c, a, b), data = dados2, start = guess, trace =
TRUE) :
fator de passos 0.000488281 reduzido abaixo de 'minFactor' de 0.000976562
>
Muito Obrigada,
Michelle
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://listas.inf.ufpr.br/pipermail/r-br/attachments/20151008/7e6f4f3d/attachment.html>
Mais detalhes sobre a lista de discussão R-br