Bootstraped simulations for multiblock methods
randboot.multiblock.RdFunction to perform bootstraped simulations for multiblock principal component analysis with instrumental variables or multiblock partial least squares, in order to get confidence intervals for some parameters, i.e., regression coefficients, variable and block importances
Usage
# S3 method for class 'multiblock'
randboot(object, nrepet = 199, optdim, ...)References
Carpenter, J. and Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians.Statistics in medicine, 19, 1141-1164.
Bougeard, S. and Dray S. (2018) Supervised Multiblock Analysis in R with the ade4 Package. Journal of Statistical Software, 86 (1), 1-17. doi:10.18637/jss.v086.i01
Author
Stéphanie Bougeard (stephanie.bougeard@anses.fr) and Stéphane Dray (stephane.dray@univ-lyon1.fr)
Examples
data(chickenk)
Mortality <- chickenk[[1]]
dudiY.chick <- dudi.pca(Mortality, center = TRUE, scale = TRUE, scannf =
FALSE)
ktabX.chick <- ktab.list.df(chickenk[2:5])
resmbpcaiv.chick <- mbpcaiv(dudiY.chick, ktabX.chick, scale = TRUE,
option = "uniform", scannf = FALSE, nf = 4)
## nrepet should be higher for a real analysis
test <- randboot(resmbpcaiv.chick, optdim = 4, nrepet = 10)
test
#> $XYcoef
#> $XYcoef$Mort7
#> Multiple bootstrap
#> Call: randboot.multiblock(object = resmbpcaiv.chick, nrepet = 10, optdim = 4)
#>
#> Number of statistics: 20
#>
#> Confidence Interval:
#> N.rep Obs 2.5% 97.5%
#> Area 10 0.085950032 -0.11722961 0.1927070289
#> Soak 10 -0.127497042 -0.24761683 0.0085047096
#> Heat 10 -0.265489511 -0.33843260 -0.1163922342
#> Sort 10 0.211732110 0.02803355 0.3439664612
#> Renov 10 0.123235977 -0.05683740 0.1939074173
#> Vitmin 10 0.029172129 -0.37357592 0.2123981714
#> Freqchick 10 -0.059863998 -0.14571432 0.0258490843
#> Homochick 10 -0.516490179 -0.68473814 -0.1950193115
#> NbChick 10 -0.145755039 -0.41769642 0.0150583449
#> Typrod 10 0.137730758 0.03972340 0.3264639909
#> Homochicken 10 -0.214346185 -0.41825622 -0.0996894679
#> Strain 10 -0.084976485 -0.19141665 0.1692919098
#> Locpb 10 -0.150139755 -0.23600021 -0.0242557121
#> Stress 10 -0.042997762 -0.25068532 0.1575087806
#> Freqchicken 10 -0.013488658 -0.16296898 0.0408429174
#> LoadType 10 -0.013287061 -0.05613712 0.1244679568
#> RainWind 10 -0.001111215 -0.12101210 0.1053167321
#> StockingD 10 0.021557816 -0.19755291 0.2355405013
#> Dlairage 10 -0.169175517 -0.20503781 0.0006203248
#> Evisc 10 -0.087462628 -0.30865263 0.1634811677
#>
#> $XYcoef$Mort
#> Multiple bootstrap
#> Call: randboot.multiblock(object = resmbpcaiv.chick, nrepet = 10, optdim = 4)
#>
#> Number of statistics: 20
#>
#> Confidence Interval:
#> N.rep Obs 2.5% 97.5%
#> Area 10 -0.02177842 -0.06993277 0.073456766
#> Soak 10 -0.16676037 -0.26003319 -0.077689425
#> Heat 10 -0.19233216 -0.35857596 -0.102742022
#> Sort 10 0.07785588 -0.00861465 0.204223142
#> Renov 10 0.03201311 -0.10727252 0.132442969
#> Vitmin 10 -0.10167257 -0.18670597 -0.073456506
#> Freqchick 10 0.01497534 -0.03298545 0.148229120
#> Homochick 10 -0.11777054 -0.27163774 -0.001105470
#> NbChick 10 -0.03508333 -0.09012902 0.046046019
#> Typrod 10 0.34305114 0.10958300 0.576485539
#> Homochicken 10 -0.26393644 -0.45390777 -0.126109292
#> Strain 10 -0.52341446 -0.66618523 -0.297833959
#> Locpb 10 0.41893706 0.28790499 0.592092485
#> Stress 10 0.77012837 0.51541383 1.111112175
#> Freqchicken 10 0.02453140 -0.08640149 0.212674230
#> LoadType 10 0.04633719 0.04792984 0.148410329
#> RainWind 10 0.07555523 0.02517507 0.143698470
#> StockingD 10 0.13001678 0.03984566 0.204386706
#> Dlairage 10 0.03741639 -0.06491490 0.161995690
#> Evisc 10 -0.04820679 -0.16862624 -0.001776209
#>
#> $XYcoef$Doa
#> Multiple bootstrap
#> Call: randboot.multiblock(object = resmbpcaiv.chick, nrepet = 10, optdim = 4)
#>
#> Number of statistics: 20
#>
#> Confidence Interval:
#> N.rep Obs 2.5% 97.5%
#> Area 10 0.224056197 0.149016208 0.32922230
#> Soak 10 -0.264513064 -0.412727200 -0.18347536
#> Heat 10 0.011755267 -0.150572571 0.22487262
#> Sort 10 0.121222302 0.097681639 0.29470539
#> Renov 10 0.002419221 -0.091100443 0.11397251
#> Vitmin 10 -0.073526898 -0.208002137 -0.03248789
#> Freqchick 10 -0.016030653 -0.086379207 0.11514994
#> Homochick 10 -0.069192224 -0.186290873 -0.04250750
#> NbChick 10 0.182877961 0.059270238 0.29958503
#> Typrod 10 0.025942608 -0.217063001 0.24743178
#> Homochicken 10 0.148044278 0.144442664 0.32896066
#> Strain 10 -0.189459932 -0.354135204 -0.01760130
#> Locpb 10 0.189734967 -0.019595178 0.39767082
#> Stress 10 0.167973023 0.003441463 0.30324331
#> Freqchicken 10 0.092561748 -0.014540637 0.25776584
#> LoadType 10 0.707231857 0.581824982 1.03012157
#> RainWind 10 0.318118255 0.101638365 0.44560732
#> StockingD 10 0.422590200 0.293265005 0.74636390
#> Dlairage 10 0.294454715 0.218396276 0.46151879
#> Evisc 10 -0.180765391 -0.363776683 0.08313042
#>
#> $XYcoef$Condemn
#> Multiple bootstrap
#> Call: randboot.multiblock(object = resmbpcaiv.chick, nrepet = 10, optdim = 4)
#>
#> Number of statistics: 20
#>
#> Confidence Interval:
#> N.rep Obs 2.5% 97.5%
#> Area 10 0.27006287 0.142178121 0.41099136
#> Soak 10 -0.22808090 -0.305379512 -0.10804417
#> Heat 10 -0.09002826 -0.158979053 0.07049358
#> Sort 10 0.15250842 -0.004301207 0.36815647
#> Renov 10 0.18942542 0.078670263 0.33838920
#> Vitmin 10 -0.12573229 -0.144931754 -0.06023857
#> Freqchick 10 -0.20924391 -0.264845824 -0.10942066
#> Homochick 10 -0.04480514 -0.191409276 0.03295749
#> NbChick 10 0.18294103 0.102974690 0.35334923
#> Typrod 10 -0.23231059 -0.321175229 -0.10580676
#> Homochicken 10 -0.28426266 -0.438070163 -0.05821824
#> Strain 10 -0.31450938 -0.551959481 -0.10438411
#> Locpb 10 0.31445584 0.058164156 0.60797832
#> Stress 10 0.33424825 0.276800667 0.57172628
#> Freqchicken 10 -0.21904521 -0.328319066 -0.13703450
#> LoadType 10 -0.12741227 -0.253438110 -0.07382482
#> RainWind 10 -0.03286434 -0.207753119 0.03141304
#> StockingD 10 -0.19985852 -0.333404373 -0.06223455
#> Dlairage 10 -0.20701408 -0.329274163 -0.15961795
#> Evisc 10 0.49570852 0.370512287 0.67654246
#>
#>
#> $bipc
#> Multiple bootstrap
#> Call: randboot.multiblock(object = resmbpcaiv.chick, nrepet = 10, optdim = 4)
#>
#> Number of statistics: 4
#>
#> Confidence Interval:
#> N.rep Obs 2.5% 97.5%
#> FarmStructure 10 0.1396215 0.06054990 0.1521305
#> OnFarmHistory 10 0.1699454 0.08229387 0.2219034
#> FlockCharacteristics 10 0.4130085 0.34383467 0.5327445
#> CatchingTranspSlaught 10 0.2774246 0.23000541 0.3389283
#>
#> $vipc
#> Multiple bootstrap
#> Call: randboot.multiblock(object = resmbpcaiv.chick, nrepet = 10, optdim = 4)
#>
#> Number of statistics: 20
#>
#> Confidence Interval:
#> N.rep Obs 2.5% 97.5%
#> Area 10 0.019021226 -0.0487693158 0.031300652
#> Soak 10 0.021212869 0.0045572563 0.031483591
#> Heat 10 0.013197420 -0.0192766026 0.017495712
#> Sort 10 0.013099040 -0.0318714930 0.024684234
#> Renov 10 0.007421402 -0.0228307785 0.011388496
#> Vitmin 10 0.003391625 -0.0239807297 0.003601814
#> Freqchick 10 0.004753272 -0.0174722801 0.003497079
#> Homochick 10 0.064052395 -0.0182858568 0.109744169
#> NbChick 10 0.015960399 -0.0021830789 0.027440544
#> Typrod 10 0.052742131 0.0153266703 0.077651196
#> Homochicken 10 0.055354532 0.0207032402 0.101854376
#> Strain 10 0.112544049 0.0581767749 0.175858893
#> Locpb 10 0.087275951 0.0033359999 0.144357715
#> Stress 10 0.193484920 0.0340159353 0.311277074
#> Freqchicken 10 0.018566556 0.0001286926 0.028372051
#> LoadType 10 0.146563153 0.1331028436 0.243338749
#> RainWind 10 0.027262859 -0.0255307792 0.044066503
#> StockingD 10 0.052194126 0.0202164390 0.097902540
#> Dlairage 10 0.043227691 0.0286235647 0.068310326
#> Evisc 10 0.048674385 -0.0276145273 0.079497708
#>
if(adegraphicsLoaded())
plot(test$bipc)