I am performing a Newey-West test to assess an estimator of a regression with heteroskedastic and autocorrelated residuals.
I am using the "sandwich" and "lmtest" packages.
While I can easily reproduce examples found on other sites, my own script causes the error:
Error in dimnames(cd) <- list(as.character(index(x)), colnames(x)) : 'dimnames' applied to non-array
My code:
ffregression <- lm(ex.return ~ fff$Mkt.RF + fff$SMB + fff$HML)
coeftest(ffregression,vcov=NeweyWest)
str(ffregression):
List of 12
$ coefficients : Named num [1:4] 0.00604 0.72976 0.90351 0.13548
..- attr(*, "names")= chr [1:4] "(Intercept)" "fff$Mkt.RF" "fff$SMB" "fff$HML"
$ residuals :An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34] -0.03637 0.0408 -0.00672 0.04648 -0.02275 ...
Indexed by objects of class: [yearmon] TZ:
Original class: 'double'
xts Attributes:
NULL
$ effects :An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34] -0.0606 0.1785 0.1379 0.0204 -0.0262 ...
Indexed by objects of class: [yearmon] TZ:
Original class: 'double'
xts Attributes:
NULL
$ rank : int 4
$ fitted.values:An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34] -0.000725 -0.031716 0.003868 0.051386 -0.047005 ...
Indexed by objects of class: [yearmon] TZ:
Original class: 'double'
xts Attributes:
NULL
$ assign : int [1:4] 0 1 2 3
$ qr :List of 5
..$ qr : num [1:34, 1:4] -5.831 0.171 0.171 0.171 0.171 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:34] "1" "2" "3" "4" ...
.. .. ..$ : chr [1:4] "(Intercept)" "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. ..- attr(*, "assign")= int [1:4] 0 1 2 3
..$ qraux: num [1:4] 1.17 1.05 1.14 1.14
..$ pivot: int [1:4] 1 2 3 4
..$ tol : num 1e-07
..$ rank : int 4
..- attr(*, "class")= chr "qr"
$ df.residual : int 30
$ xlevels : Named list()
$ call : language lm(formula = ex.return ~ fff$Mkt.RF + fff$SMB + fff$HML)
$ terms :Classes 'terms', 'formula' language ex.return ~ fff$Mkt.RF + fff$SMB + fff$HML
.. ..- attr(*, "variables")= language list(ex.return, fff$Mkt.RF, fff$SMB, fff$HML)
.. ..- attr(*, "factors")= int [1:4, 1:3] 0 1 0 0 0 0 1 0 0 0 ...
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:4] "ex.return" "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. .. .. ..$ : chr [1:3] "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. ..- attr(*, "term.labels")= chr [1:3] "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. ..- attr(*, "order")= int [1:3] 1 1 1
.. ..- attr(*, "intercept")= int 1
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. ..- attr(*, "predvars")= language list(ex.return, fff$Mkt.RF, fff$SMB, fff$HML)
.. ..- attr(*, "dataClasses")= Named chr [1:4] "nmatrix.1" "nmatrix.1" "nmatrix.1" "nmatrix.1"
.. .. ..- attr(*, "names")= chr [1:4] "ex.return" "fff$Mkt.RF" "fff$SMB" "fff$HML"
$ model :'data.frame': 34 obs. of 4 variables:
..$ ex.return :An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34, 1] -0.03709 0.00909 -0.00285 0.09786 -0.06975 ...
- attr(*, "dimnames")=List of 2
..$ : NULL
..$ : chr "RF"
Indexed by objects of class: [yearmon] TZ:
Original class: 'double'
xts Attributes:
NULL
..$ fff$Mkt.RF:An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34, 1] 0.0043 -0.0019 0.0206 0.0261 -0.0204 0.0424 -0.0197 0.0252 0.0255 -0.0006 ...
- attr(*, "dimnames")=List of 2
..$ : NULL
..$ : chr "Mkt.RF"
Indexed by objects of class: [yearmon] TZ:
xts Attributes:
NULL
..$ fff$SMB :An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34, 1] -0.0185 -0.0419 -0.0185 0.0301 -0.0422 0.004 -0.038 0.0428 -0.0205 0.0259 ...
- attr(*, "dimnames")=List of 2
..$ : NULL
..$ : chr "SMB"
Indexed by objects of class: [yearmon] TZ:
xts Attributes:
NULL
..$ fff$HML :An ‘xts’ object on Mar 2014/Dec 2016 containing:
Data: num [1:34, 1] 0.0503 0.011 -0.0036 -0.0066 -0.0002 -0.0055 -0.0119 -0.0168 -0.0298 0.0212 ...
- attr(*, "dimnames")=List of 2
..$ : NULL
..$ : chr "HML"
Indexed by objects of class: [yearmon] TZ:
xts Attributes:
NULL
..- attr(*, "terms")=Classes 'terms', 'formula' language ex.return ~ fff$Mkt.RF + fff$SMB + fff$HML
.. .. ..- attr(*, "variables")= language list(ex.return, fff$Mkt.RF, fff$SMB, fff$HML)
.. .. ..- attr(*, "factors")= int [1:4, 1:3] 0 1 0 0 0 0 1 0 0 0 ...
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:4] "ex.return" "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. .. .. .. ..$ : chr [1:3] "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. .. ..- attr(*, "term.labels")= chr [1:3] "fff$Mkt.RF" "fff$SMB" "fff$HML"
.. .. ..- attr(*, "order")= int [1:3] 1 1 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. .. ..- attr(*, "predvars")= language list(ex.return, fff$Mkt.RF, fff$SMB, fff$HML)
.. .. ..- attr(*, "dataClasses")= Named chr [1:4] "nmatrix.1" "nmatrix.1" "nmatrix.1" "nmatrix.1"
.. .. .. ..- attr(*, "names")= chr [1:4] "ex.return" "fff$Mkt.RF" "fff$SMB" "fff$HML"
- attr(*, "class")= chr "lm"
Example:
set.seed(04012017)
n<-34
correlated_residuals<-arima.sim(list(ar = .9), n)
y<-correlated_residuals
x<-1:n
plot(x,correlated_residuals)
fit<-lm(y~x)
abline(fit)
summary(fit) # standard estimates
coeftest(fit,vcov=NeweyWest(fit,verbose=T))
str(fit):
List of 12
$ coefficients : Named num [1:2] -0.179 0.148
..- attr(*, "names")= chr [1:2] "(Intercept)" "x"
$ residuals : Named num [1:34] -0.9529 0.976 0.3025 -0.0486 -1.1214 ...
..- attr(*, "names")= chr [1:34] "1" "2" "3" "4" ...
$ effects : Named num [1:34] -14.026 8.449 0.25 -0.085 -1.142 ...
..- attr(*, "names")= chr [1:34] "(Intercept)" "x" "" "" ...
$ rank : int 2
$ fitted.values: Named num [1:34] -0.0314 0.1163 0.264 0.4116 0.5593 ...
..- attr(*, "names")= chr [1:34] "1" "2" "3" "4" ...
$ assign : int [1:2] 0 1
$ qr :List of 5
..$ qr : num [1:34, 1:2] -5.831 0.171 0.171 0.171 0.171 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:34] "1" "2" "3" "4" ...
.. .. ..$ : chr [1:2] "(Intercept)" "x"
.. ..- attr(*, "assign")= int [1:2] 0 1
..$ qraux: num [1:2] 1.17 1.23
..$ pivot: int [1:2] 1 2
..$ tol : num 1e-07
..$ rank : int 2
..- attr(*, "class")= chr "qr"
$ df.residual : int 32
$ xlevels : Named list()
$ call : language lm(formula = y ~ x)
$ terms :Classes 'terms', 'formula' language y ~ x
.. ..- attr(*, "variables")= language list(y, x)
.. ..- attr(*, "factors")= int [1:2, 1] 0 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:2] "y" "x"
.. .. .. ..$ : chr "x"
.. ..- attr(*, "term.labels")= chr "x"
.. ..- attr(*, "order")= int 1
.. ..- attr(*, "intercept")= int 1
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. ..- attr(*, "predvars")= language list(y, x)
.. ..- attr(*, "dataClasses")= Named chr [1:2] "numeric" "numeric"
.. .. ..- attr(*, "names")= chr [1:2] "y" "x"
$ model :'data.frame': 34 obs. of 2 variables:
..$ y: num [1:34] -0.984 1.092 0.566 0.363 -0.562 ...
..$ x: int [1:34] 1 2 3 4 5 6 7 8 9 10 ...
..- attr(*, "terms")=Classes 'terms', 'formula' language y ~ x
.. .. ..- attr(*, "variables")= language list(y, x)
.. .. ..- attr(*, "factors")= int [1:2, 1] 0 1
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:2] "y" "x"
.. .. .. .. ..$ : chr "x"
.. .. ..- attr(*, "term.labels")= chr "x"
.. .. ..- attr(*, "order")= int 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. .. ..- attr(*, "predvars")= language list(y, x)
.. .. ..- attr(*, "dataClasses")= Named chr [1:2] "numeric" "numeric"
.. .. .. ..- attr(*, "names")= chr [1:2] "y" "x"
- attr(*, "class")= chr "lm"
It seems that R is complaining that it cannot assign dimnames because your data is coming from a non-array. Try to use only one xts object for your data instead.
fff$ex.return <- ex.return
ffregression <- lm(ex.return ~ Mkt.RF + SMB + HML, data = fff)
coeftest(ffregression,vcov=NeweyWest)
Related
I have a list called Transitions that has results for 3 size groups: Small, Medium, and Large. In each size group are multiple matrices. The nested element I'm interested in is Transitions$"Size Group"$nij.
I would like to write a function that takes all three size elements and their respective nij matrix elements and adds them together.
Something similar to Reduce('+', List of Matrices)
I manually collected the nij matrices and made a list to show what I have.
Transitions.ls <-
list(Small = structure(c(38L, 83L, 112L, 58L, 142L, 83L, 124L,
62L, 23L), .Dim = c(3L, 3L), .Dimnames = list(i = c("Downgraded",
"Unchanged", "Upgraded"), j = c("Downgraded", "Unchanged", "Upgraded"
)), class = "table"), Medium = structure(c(83L, 184L, 210L, 134L,
243L, 183L, 256L, 122L, 86L), .Dim = c(3L, 3L), .Dimnames = list(
i = c("Downgraded", "Unchanged", "Upgraded"), j = c("Downgraded",
"Unchanged", "Upgraded")), class = "table"), Large = structure(c(3L,
9L, 11L, 6L, 4L, 18L, 21L, 13L, 10L), .Dim = c(3L, 3L), .Dimnames = list(
i = c("Downgraded", "Unchanged", "Upgraded"), j = c("Downgraded",
"Unchanged", "Upgraded")), class = "table"))
Each matrix in the list above was nested. I did the following to obtain them:
Transitions.ls <- list(Transitions$Small$nij, Transitions$Medium$nij,Transitions$Large$nij)
My desired output is:
Reduce('+', Transitions.ls)
j
i Downgraded Unchanged Upgraded
Downgraded 124 198 401
Unchanged 276 389 197
Upgraded 333 284 119
Data structure of the original list Transitions:
List of 3
$ Small :List of 12
..$ nij : 'table' int [1:3, 1:3] 38 83 112 58 142 83 124 62 23
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ ni : Named num [1:3] 220 287 218
.. ..- attr(*, "names")= chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ pij : 'table' num [1:3, 1:3] 0.173 0.289 0.514 0.264 0.495 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ pj : Named num [1:3] 0.321 0.39 0.288
.. ..- attr(*, "names")= chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ se.pij : 'table' num [1:3, 1:3] 0.0255 0.0268 0.0339 0.0297 0.0295 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ nij.out : 'xtabs' int [1:3, 1:3] 1 3 3 1 8 4 2 1 1
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
..$ ni.out : Named num [1:3] 4 12 8
.. ..- attr(*, "names")= chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ pij.out : 'xtabs' num [1:3, 1:3] 0.25 0.25 0.375 0.25 0.667 ...
.. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ pj.out : Named num [1:3] 0.292 0.542 0.167
.. ..- attr(*, "names")= chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ se.pij.out : 'xtabs' num [1:3, 1:3] 0.217 0.125 0.171 0.217 0.136 ...
.. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ p.diff : 'table' num [1:3, 1:3] -0.0773 0.0392 0.1388 0.0136 -0.1719 ...
.. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
..$ Transitions:List of 16
.. ..$ 1998: 'xtabs' int [1:3, 1:3] 2 10 12 5 11 6 7 6 2
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 1999: 'xtabs' int [1:3, 1:3] 4 9 8 6 8 5 14 6 2
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2000: 'xtabs' int [1:3, 1:3] 5 8 12 6 10 8 10 4 4
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2001: 'xtabs' int [1:3, 1:3] 6 5 10 8 12 5 9 7 3
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2002: 'xtabs' int [1:3, 1:3] 4 8 10 6 12 7 10 7 3
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2003: 'xtabs' int [1:3, 1:3] 3 8 12 4 6 7 13 6 0
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2004: 'xtabs' int [1:3, 1:3] 1 4 7 5 5 10 12 6 0
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2005: 'xtabs' int [1:3, 1:3] 1 5 6 3 11 8 8 3 2
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2006: 'xtabs' int [1:3, 1:3] 4 7 7 1 13 4 6 2 2
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2007: 'xtabs' int [1:3, 1:3] 3 2 4 4 10 3 9 2 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2008: 'xtabs' int [1:3, 1:3] 1 3 3 2 8 5 3 2 0
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2009: 'xtabs' int [1:3, 1:3] 1 5 4 2 5 0 4 5 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2010: 'xtabs' int [1:3, 1:3] 1 2 8 4 4 4 6 1 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2011: 'xtabs' int [1:3, 1:3] 1 3 3 1 9 4 6 2 0
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2012: 'xtabs' int [1:3, 1:3] 0 1 3 0 10 3 5 2 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
.. ..$ 2013: 'xtabs' int [1:3, 1:3] 1 3 3 1 8 4 2 1 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ i: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. .. ..$ j: chr [1:3] "Downgraded" "Unchanged" "Upgraded"
.. .. ..- attr(*, "call")= language xtabs(formula = n ~ i + j, data = .)
Perhaps, we can loop over the "Size Group" vector, extract the elements with [[, and then extract the 'nij' (either $ or [[ can be used here), then apply Reduce
v1 <- c("Small", "Medium", "Large")
Reduce(`+`, lapply(v1, function(nm) Transitions[[nm]]$nij))
I have a data frame APD which consists of a factor (Serial_number) and each element of this factor has some observables (Amplification against Voltage). Now I would like to apply the function invest from the investr package to do an inverse regression across all serial numbers (means an inverse regression separated for each serial number).
This works very well if I only use the dataset of one Serial_number, e.g.:
> model<- lm(Amplification ~ Voltage + I(Voltage^2), data = APD[APD$Serial_number==608004648,])
> inverse<- invest(model, y0=150, interval=c("inversion", "Wald", "percentile", "none"),
level=0.95, mean.response = FALSE
)
> inverse
estimate lower upper
385.4670 385.0665 385.8446
When I take the whole data frame:
> models<- dlply(APD, "Serial_number", function(df) lm(Amplification ~ Voltage + I(Voltage^2), data = df))
> inverse<- invest(models, y0=150, interval=c("inversion", "Wald", "percentile", "none"),
level=0.95, mean.response = FALSE
)
> Error in UseMethod("invest") :
no applicable method for 'invest' applied to an object of class "list"
I tried inverse<- invest(models[1], y0=150, interval=c("inversion", "Wald", "percentile", "none") but it's the same problem.
I can't get my head around how to proceed further. I think this must be such a daily issue there are probably already practical solutions present. Don't they?
My next step would be to create a lm-object for each Serial_number but also this seems unnecessary complicated.
> model<- lm(Amplification ~ Voltage + I(Voltage^2), data = APD[APD$Serial_number==608004648,])
> str(model)
List of 12
$ coefficients : Named num [1:3] 59714.045 -318.043 0.424
..- attr(*, "names")= chr [1:3] "(Intercept)" "Voltage" "I(Voltage^2)"
$ residuals : Named num [1:4] -0.0483 0.1458 -0.1459 0.0484
..- attr(*, "names")= chr [1:4] "1" "2" "3" "4"
$ effects : Named num [1:4] -295.61 -38.412 3.412 -0.217
..- attr(*, "names")= chr [1:4] "(Intercept)" "Voltage" "I(Voltage^2)" ""
$ rank : int 3
$ fitted.values: Named num [1:4] 124 138 155 175
..- attr(*, "names")= chr [1:4] "1" "2" "3" "4"
$ assign : int [1:3] 0 1 2
$ qr :List of 5
..$ qr : num [1:4, 1:3] -2 0.5 0.5 0.5 -770 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:4] "1" "2" "3" "4"
.. .. ..$ : chr [1:3] "(Intercept)" "Voltage" "I(Voltage^2)"
.. ..- attr(*, "assign")= int [1:3] 0 1 2
..$ qraux: num [1:3] 1.5 1 1.37
..$ pivot: int [1:3] 1 2 3
..$ tol : num 1e-07
..$ rank : int 3
..- attr(*, "class")= chr "qr"
$ df.residual : int 1
$ xlevels : Named list()
$ call : language lm(formula = Amplification ~ Voltage + I(Voltage^2), data = APD[APD$Serial_number == 608004648, ])
$ terms :Classes 'terms', 'formula' language Amplification ~ Voltage + I(Voltage^2)
.. ..- attr(*, "variables")= language list(Amplification, Voltage, I(Voltage^2))
.. ..- attr(*, "factors")= int [1:3, 1:2] 0 1 0 0 0 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
.. .. .. ..$ : chr [1:2] "Voltage" "I(Voltage^2)"
.. ..- attr(*, "term.labels")= chr [1:2] "Voltage" "I(Voltage^2)"
.. ..- attr(*, "order")= int [1:2] 1 1
.. ..- attr(*, "intercept")= int 1
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. ..- attr(*, "predvars")= language list(Amplification, Voltage, I(Voltage^2))
.. ..- attr(*, "dataClasses")= Named chr [1:3] "numeric" "numeric" "numeric"
.. .. ..- attr(*, "names")= chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
$ model :'data.frame': 4 obs. of 3 variables:
..$ Amplification: num [1:4] 124 138 155 175
..$ Voltage : num [1:4] 382 384 386 388
..$ I(Voltage^2) :Class 'AsIs' num [1:4] 145900 147440 148975 150528
..- attr(*, "terms")=Classes 'terms', 'formula' language Amplification ~ Voltage + I(Voltage^2)
.. .. ..- attr(*, "variables")= language list(Amplification, Voltage, I(Voltage^2))
.. .. ..- attr(*, "factors")= int [1:3, 1:2] 0 1 0 0 0 1
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
.. .. .. .. ..$ : chr [1:2] "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "term.labels")= chr [1:2] "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "order")= int [1:2] 1 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. .. ..- attr(*, "predvars")= language list(Amplification, Voltage, I(Voltage^2))
.. .. ..- attr(*, "dataClasses")= Named chr [1:3] "numeric" "numeric" "numeric"
.. .. .. ..- attr(*, "names")= chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
- attr(*, "class")= chr "lm"
str(models) are hundreds of lines.. a few are:
models<- dlply(APD, "Serial_number", function(df) lm(Amplification ~ Voltage + I(Voltage^2), data = df))
str(models)
(...)
$ 713006445 :List of 12
..$ coefficients : Named num [1:3] 68847.991 -386.479 0.543
.. ..- attr(*, "names")= chr [1:3] "(Intercept)" "Voltage" "I(Voltage^2)"
..$ residuals : Named num [1:4] -0.0655 0.1964 -0.1962 0.0653
.. ..- attr(*, "names")= chr [1:4] "1" "2" "3" "4"
..$ effects : Named num [1:4] -317.503 45.226 4.347 0.293
.. ..- attr(*, "names")= chr [1:4] "(Intercept)" "Voltage" "I(Voltage^2)" ""
..$ rank : int 3
..$ fitted.values: Named num [1:4] 131 146 167 191
.. ..- attr(*, "names")= chr [1:4] "1" "2" "3" "4"
..$ assign : int [1:3] 0 1 2
..$ qr :List of 5
.. ..$ qr : num [1:4, 1:3] -2 0.5 0.5 0.5 -730.1 ...
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:4] "1" "2" "3" "4"
.. .. .. ..$ : chr [1:3] "(Intercept)" "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "assign")= int [1:3] 0 1 2
.. ..$ qraux: num [1:3] 1.5 1 1.96
.. ..$ pivot: int [1:3] 1 2 3
.. ..$ tol : num 1e-07
.. ..$ rank : int 3
.. ..- attr(*, "class")= chr "qr"
..$ df.residual : int 1
..$ xlevels : Named list()
..$ call : language lm(formula = Amplification ~ Voltage + I(Voltage^2), data = df)
..$ terms :Classes 'terms', 'formula' language Amplification ~ Voltage + I(Voltage^2)
.. .. ..- attr(*, "variables")= language list(Amplification, Voltage, I(Voltage^2))
.. .. ..- attr(*, "factors")= int [1:3, 1:2] 0 1 0 0 0 1
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
.. .. .. .. ..$ : chr [1:2] "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "term.labels")= chr [1:2] "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "order")= int [1:2] 1 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: 0xb810100>
.. .. ..- attr(*, "predvars")= language list(Amplification, Voltage, I(Voltage^2))
.. .. ..- attr(*, "dataClasses")= Named chr [1:3] "numeric" "numeric" "numeric"
.. .. .. ..- attr(*, "names")= chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
..$ model :'data.frame': 4 obs. of 3 variables:
.. ..$ Amplification: num [1:4] 131 147 166 191
.. ..$ Voltage : num [1:4] 362 364 366 368
.. ..$ I(Voltage^2) :Class 'AsIs' num [1:4] 131064 132516 133974 135445
.. ..- attr(*, "terms")=Classes 'terms', 'formula' language Amplification ~ Voltage + I(Voltage^2)
.. .. .. ..- attr(*, "variables")= language list(Amplification, Voltage, I(Voltage^2))
.. .. .. ..- attr(*, "factors")= int [1:3, 1:2] 0 1 0 0 0 1
.. .. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. .. ..$ : chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
.. .. .. .. .. ..$ : chr [1:2] "Voltage" "I(Voltage^2)"
.. .. .. ..- attr(*, "term.labels")= chr [1:2] "Voltage" "I(Voltage^2)"
.. .. .. ..- attr(*, "order")= int [1:2] 1 1
.. .. .. ..- attr(*, "intercept")= int 1
.. .. .. ..- attr(*, "response")= int 1
.. .. .. ..- attr(*, ".Environment")=<environment: 0xb810100>
.. .. .. ..- attr(*, "predvars")= language list(Amplification, Voltage, I(Voltage^2))
.. .. .. ..- attr(*, "dataClasses")= Named chr [1:3] "numeric" "numeric" "numeric"
.. .. .. .. ..- attr(*, "names")= chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
..- attr(*, "class")= chr "lm"
$ 713006446 :List of 12
..$ coefficients : Named num [1:3] 58059.34 -331.908 0.475
.. ..- attr(*, "names")= chr [1:3] "(Intercept)" "Voltage" "I(Voltage^2)"
..$ residuals : Named num [1:4] -0.0574 0.1729 -0.1727 0.0573
.. ..- attr(*, "names")= chr [1:4] "1" "2" "3" "4"
..$ effects : Named num [1:4] -300.53 -41.391 3.808 -0.258
.. ..- attr(*, "names")= chr [1:4] "(Intercept)" "Voltage" "I(Voltage^2)" ""
..$ rank : int 3
..$ fitted.values: Named num [1:4] 124 139 158 180
.. ..- attr(*, "names")= chr [1:4] "1" "2" "3" "4"
..$ assign : int [1:3] 0 1 2
..$ qr :List of 5
.. ..$ qr : num [1:4, 1:3] -2 0.5 0.5 0.5 -717.9 ...
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:4] "1" "2" "3" "4"
.. .. .. ..$ : chr [1:3] "(Intercept)" "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "assign")= int [1:3] 0 1 2
.. ..$ qraux: num [1:3] 1.5 1 1.37
.. ..$ pivot: int [1:3] 1 2 3
.. ..$ tol : num 1e-07
.. ..$ rank : int 3
.. ..- attr(*, "class")= chr "qr"
..$ df.residual : int 1
..$ xlevels : Named list()
..$ call : language lm(formula = Amplification ~ Voltage + I(Voltage^2), data = df)
..$ terms :Classes 'terms', 'formula' language Amplification ~ Voltage + I(Voltage^2)
.. .. ..- attr(*, "variables")= language list(Amplification, Voltage, I(Voltage^2))
.. .. ..- attr(*, "factors")= int [1:3, 1:2] 0 1 0 0 0 1
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
.. .. .. .. ..$ : chr [1:2] "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "term.labels")= chr [1:2] "Voltage" "I(Voltage^2)"
.. .. ..- attr(*, "order")= int [1:2] 1 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: 0xb6fc150>
.. .. ..- attr(*, "predvars")= language list(Amplification, Voltage, I(Voltage^2))
.. .. ..- attr(*, "dataClasses")= Named chr [1:3] "numeric" "numeric" "numeric"
.. .. .. ..- attr(*, "names")= chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
..$ model :'data.frame': 4 obs. of 3 variables:
.. ..$ Amplification: num [1:4] 124 139 157 180
.. ..$ Voltage : num [1:4] 356 358 360 362
.. ..$ I(Voltage^2) :Class 'AsIs' num [1:4] 126715 128145 129576 131024
.. ..- attr(*, "terms")=Classes 'terms', 'formula' language Amplification ~ Voltage + I(Voltage^2)
.. .. .. ..- attr(*, "variables")= language list(Amplification, Voltage, I(Voltage^2))
.. .. .. ..- attr(*, "factors")= int [1:3, 1:2] 0 1 0 0 0 1
.. .. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. .. ..$ : chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
.. .. .. .. .. ..$ : chr [1:2] "Voltage" "I(Voltage^2)"
.. .. .. ..- attr(*, "term.labels")= chr [1:2] "Voltage" "I(Voltage^2)"
.. .. .. ..- attr(*, "order")= int [1:2] 1 1
.. .. .. ..- attr(*, "intercept")= int 1
.. .. .. ..- attr(*, "response")= int 1
.. .. .. ..- attr(*, ".Environment")=<environment: 0xb6fc150>
.. .. .. ..- attr(*, "predvars")= language list(Amplification, Voltage, I(Voltage^2))
.. .. .. ..- attr(*, "dataClasses")= Named chr [1:3] "numeric" "numeric" "numeric"
.. .. .. .. ..- attr(*, "names")= chr [1:3] "Amplification" "Voltage" "I(Voltage^2)"
..- attr(*, "class")= chr "lm"
$ 713006449 :List of 12
..$ coefficients : Named num [1:3] 56708.406 -324.219 0.464
(...)
I use RStudio/Ubuntu 14.04.
edit: Due to advices of joran and Gregor I tried the following:
> invest(models[[1]], y0=150, interval=c("inversion", "Wald", "percentile", "none"), level=0.95, mean.response = FALSE)
Error in invest.lm(models[[1]], y0 = 150, interval = c("inversion", "Wald", :
'x0.name' is missing, please select a valid predictor variable
and
in case of lapply I'm not sure how to define it > lapply(X, FUN, ...) and in my case this would be lapply(object, invest(object, ..)) :
> lapply(models, invest(models, y0=150, interval=c("inversion", "Wald", "percentile", "none"), level=0.95, mean.response = FALSE))
Error in UseMethod("invest") :
no applicable method for 'invest' applied to an object of class "list"
editedit:
I tried to use a lmList as follows:
library(nlme)
models<- lmList(Amplification ~ Voltage + I(Voltage^2)|Serial_number, data= APD)
> invest(models[[1]], y0=150, interval=c("inversion", "Wald", "percentile", "none"),
+ level=0.95, mean.response = FALSE)
Error in eval(object$call$data, envir = parent.frame()) :
object 'dat' not found
Has someone an idea how to proceed? I take every solution which allows to apply a lm fit to the factor of a dataset and then to apply the invest function to each lm object.
Isn't there the possibility to extract a lm object out of a list of lm objects and it remains as an lm object?
In case someone else will have a hard time solving this "trivial" issue: I circumvented the scenario above by splitting the data frame into many data frames separated by the levels of the factor:
new_data_frame<- split(data_frame, data_frame$factor)
Next, I iterate through all data frames and fit and invest each:
for(i in 1:length(new_data_frame)){
temp_data_frame<- as.data.frame(new_data_frame[i])
fit<- lm(Response ~ Predictor, data = temp_model_frame)
inverse_regression<- invest(fit, y0=150, interval=c("inversion", "Wald", "percentile", "none"), level=0.95, mean.response = FALSE)
# some lines to store each time inverse_regression..
}
I have an object,
> str(summary(A$aov[[2]]))
List of 1
$ :Classes ‘anova’ and 'data.frame': 4 obs. of 5 variables:
..$ Df : num [1:4] 1 1 1 26
..$ Sum Sq : num [1:4] 0.00966 0.01137 0.00458 1.13068
..$ Mean Sq: num [1:4] 0.00966 0.01137 0.00458 0.04349
..$ F value: num [1:4] 0.222 0.261 0.105 NA
..$ Pr(>F) : num [1:4] 0.641 0.614 0.748 NA
- attr(*, "class")= chr [1:2] "summary.aov" "listof"
however I can't seem to figure out how to access "Sum Sq"?
I tried the following already,
summary(A$aov[[2]])$'Sum Sq'
summary(A$aov[[2]])[2]
summary(A$aov[[2]])[2,]
Would be even great for learning if I can extract the Sum Sq values without the summary function. For example, typing
> A$aov[[2]]
Terms:
Adult_Group Sex Adult_Group:Sex Residuals
Sum of Squares 0.0096577 0.0113658 0.0045836 1.1306777
Deg. of Freedom 1 1 1 26
Residual standard error: 0.2085368
Estimated effects may be unbalanced
> str(A$aov[[2]])
List of 9
$ coefficients : Named num [1:3] -0.00757 -0.01057 -0.00746
..- attr(*, "names")= chr [1:3] "Adult_Group1" "Sex1" "Adult_Group1:Sex1"
$ residuals : Named num [1:29] 0.04596 -0.00541 0.41315 -0.24305 -0.06205 ...
..- attr(*, "names")= chr [1:29] "2" "3" "4" "5" ...
$ effects : Named num [1:29] 0.0983 -0.1066 -0.0677 -0.305 -0.0284 ...
..- attr(*, "names")= chr [1:29] "Adult_Group1" "Sex1" "Adult_Group1:Sex1" "" ...
$ rank : int 3
$ fitted.values: Named num [1:29] 4.16e-17 6.51e-17 4.51e-02 3.49e-02 -1.90e-02 ...
..- attr(*, "names")= chr [1:29] "2" "3" "4" "5" ...
$ assign : int [1:3] 1 2 4
$ qr :List of 5
..$ qr : num [1:29, 1:3] -9.30 -5.97e-17 -3.23e-01 -2.50e-01 1.36e-01 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:29] "2" "3" "4" "5" ...
.. .. ..$ : chr [1:3] "Adult_Group1" "Sex1" "Adult_Group1:Sex1"
.. ..- attr(*, "assign")= int [1:3] 1 2 4
..$ qraux: num [1:3] 1 1 1.28
..$ pivot: int [1:3] 1 2 3
..$ tol : num 1e-07
..$ rank : int 3
..- attr(*, "class")= chr "qr"
$ df.residual : int 26
$ terms :Classes 'terms', 'formula' language value ~ Adult_Group + Sex + Region + Adult_Group:Sex + Adult_Group:Region + Sex:Region + Adult_Group:Sex:Region
.. ..- attr(*, "variables")= language list(value, Adult_Group, Sex, Region)
.. ..- attr(*, "factors")= int [1:4, 1:7] 0 1 0 0 0 0 1 0 0 0 ...
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:4] "value" "Adult_Group" "Sex" "Region"
.. .. .. ..$ : chr [1:7] "Adult_Group" "Sex" "Region" "Adult_Group:Sex" ...
.. ..- attr(*, "term.labels")= chr [1:7] "Adult_Group" "Sex" "Region" "Adult_Group:Sex" ...
.. ..- attr(*, "order")= int [1:7] 1 1 1 2 2 2 3
.. ..- attr(*, "intercept")= int 1
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: 0x000000005c411db8>
- attr(*, "class")= chr [1:2] "aov" "lm"
So I guess from another level, how do I get extract Sum Sq from either,
summary(A$aov[[2]]) or A$aov[[2]]
Would be great to learn how do this since str confuses me.
Assuming I have an X matrix and y vector such as the following:
X=
[,1] [,2] [,3]
[1,] 83.0 234.289 235.6 ...
[2,] 88.5 259.426 232.5 ...
[3,] 88.2 258.054 368.2 ...
y=
[1] 60.323 61.122 60.171...
After conducting a decomposition, I find the coefficients B and residuals E. How can I create a new lm object (ie lmQ) that stores my results for B and E in order to get something like this:
> lmQ(X)
$coefficients
X1 X2
B1 B2
$residuals
[1] R1 R2 R3...
Please help thanks!
Objects are just lists with a class attribute.
Create a list, assign the components, set the class -- til it looks like what str() gives.
Now, lm() returns somewhat large objects so you have some work to do:
> df <- data.frame(y=runif(10), x=rnorm(10))
> fit <- lm(y ~ x, df)
> str(fit)
List of 12
$ coefficients : Named num [1:2] 0.5368 0.0314
..- attr(*, "names")= chr [1:2] "(Intercept)" "x"
$ residuals : Named num [1:10] 0.2899 -0.3592 -0.1753 0.3187 -0.0235 ...
..- attr(*, "names")= chr [1:10] "1" "2" "3" "4" ...
$ effects : Named num [1:10] -1.734 0.104 -0.325 0.485 -0.158 ...
..- attr(*, "names")= chr [1:10] "(Intercept)" "x" "" "" ...
$ rank : int 2
$ fitted.values: Named num [1:10] 0.553 0.526 0.573 0.479 0.569 ...
..- attr(*, "names")= chr [1:10] "1" "2" "3" "4" ...
$ assign : int [1:2] 0 1
$ qr :List of 5
..$ qr : num [1:10, 1:2] -3.162 0.316 0.316 0.316 0.316 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:10] "1" "2" "3" "4" ...
.. .. ..$ : chr [1:2] "(Intercept)" "x"
.. ..- attr(*, "assign")= int [1:2] 0 1
..$ qraux: num [1:2] 1.32 1.22
..$ pivot: int [1:2] 1 2
..$ tol : num 1e-07
..$ rank : int 2
..- attr(*, "class")= chr "qr"
$ df.residual : int 8
$ xlevels : Named list()
$ call : language lm(formula = y ~ x, data = df)
$ terms :Classes 'terms', 'formula' length 3 y ~ x
.. ..- attr(*, "variables")= language list(y, x)
.. ..- attr(*, "factors")= int [1:2, 1] 0 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:2] "y" "x"
.. .. .. ..$ : chr "x"
.. ..- attr(*, "term.labels")= chr "x"
.. ..- attr(*, "order")= int 1
.. ..- attr(*, "intercept")= int 1
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. ..- attr(*, "predvars")= language list(y, x)
.. ..- attr(*, "dataClasses")= Named chr [1:2] "numeric" "numeric"
.. .. ..- attr(*, "names")= chr [1:2] "y" "x"
$ model :'data.frame': 10 obs. of 2 variables:
..$ y: num [1:10] 0.843 0.167 0.398 0.798 0.545 ...
..$ x: num [1:10] 0.504 -0.337 1.151 -1.835 1.011 ...
..- attr(*, "terms")=Classes 'terms', 'formula' length 3 y ~ x
.. .. ..- attr(*, "variables")= language list(y, x)
.. .. ..- attr(*, "factors")= int [1:2, 1] 0 1
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:2] "y" "x"
.. .. .. .. ..$ : chr "x"
.. .. ..- attr(*, "term.labels")= chr "x"
.. .. ..- attr(*, "order")= int 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. .. ..- attr(*, "predvars")= language list(y, x)
.. .. ..- attr(*, "dataClasses")= Named chr [1:2] "numeric" "numeric"
.. .. .. ..- attr(*, "names")= chr [1:2] "y" "x"
- attr(*, "class")= chr "lm"
>
I am looking to extract the p-value generated from an anova in R.
Here is what I am running:
test <- aov(asq[,9] ~ asq[,187])
summary(test)
Yields:
Df Sum Sq Mean Sq F value Pr(>F)
asq[, 187] 1 3.02 3.01951 12.333 0.0004599 ***
Residuals 1335 326.85 0.24483
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12 observations deleted due to missingness
When I look a the structure, this is what I see. I usually can work through lists to get what I need, but I am having a hard time with this one. A Google searched also seemed to reveal much simpler structures than I am getting.
NOTE: ASQ is my data frame.
str(test)
List of 13
$ coefficients : Named num [1:2] 0.2862 0.0973
..- attr(*, "names")= chr [1:2] "(Intercept)" "asq[, 187]"
$ residuals : Named num [1:1337] 0.519 0.519 -0.481 -0.481 -0.481 ...
..- attr(*, "names")= chr [1:1337] "1" "2" "3" "4" ...
$ effects : Named num [1:1337] -16.19 -1.738 -0.505 -0.505 -0.505 ...
..- attr(*, "names")= chr [1:1337] "(Intercept)" "asq[, 187]" "" "" ...
$ rank : int 2
$ fitted.values: Named num [1:1337] 0.481 0.481 0.481 0.481 0.481 ...
..- attr(*, "names")= chr [1:1337] "1" "2" "3" "4" ...
$ assign : int [1:2] 0 1
$ qr :List of 5
..$ qr : num [1:1337, 1:2] -36.565 0.0273 0.0273 0.0273 0.0273 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:1337] "1" "2" "3" "4" ...
.. .. ..$ : chr [1:2] "(Intercept)" "asq[, 187]"
.. ..- attr(*, "assign")= int [1:2] 0 1
..$ qraux: num [1:2] 1.03 1.02
..$ pivot: int [1:2] 1 2
..$ tol : num 1e-07
..$ rank : int 2
..- attr(*, "class")= chr "qr"
$ df.residual : int 1335
$ na.action :Class 'omit' Named int [1:12] 26 257 352 458 508 624 820 874 1046 1082 ...
.. ..- attr(*, "names")= chr [1:12] "26" "257" "352" "458" ...
$ xlevels : list()
$ call : language aov(formula = asq[, 9] ~ asq[, 187])
$ terms :Classes 'terms', 'formula' length 3 asq[, 9] ~ asq[, 187]
.. ..- attr(*, "variables")= language list(asq[, 9], asq[, 187])
.. ..- attr(*, "factors")= int [1:2, 1] 0 1
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:2] "asq[, 9]" "asq[, 187]"
.. .. .. ..$ : chr "asq[, 187]"
.. ..- attr(*, "term.labels")= chr "asq[, 187]"
.. ..- attr(*, "order")= int 1
.. ..- attr(*, "intercept")= int 1
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. ..- attr(*, "predvars")= language list(asq[, 9], asq[, 187])
.. ..- attr(*, "dataClasses")= Named chr [1:2] "numeric" "numeric"
.. .. ..- attr(*, "names")= chr [1:2] "asq[, 9]" "asq[, 187]"
$ model :'data.frame': 1337 obs. of 2 variables:
..$ asq[, 9] : int [1:1337] 1 1 0 0 0 1 1 1 0 0 ...
..$ asq[, 187]: int [1:1337] 2 2 2 2 2 2 2 2 2 2 ...
..- attr(*, "terms")=Classes 'terms', 'formula' length 3 asq[, 9] ~ asq[, 187]
.. .. ..- attr(*, "variables")= language list(asq[, 9], asq[, 187])
.. .. ..- attr(*, "factors")= int [1:2, 1] 0 1
.. .. .. ..- attr(*, "dimnames")=List of 2
.. .. .. .. ..$ : chr [1:2] "asq[, 9]" "asq[, 187]"
.. .. .. .. ..$ : chr "asq[, 187]"
.. .. ..- attr(*, "term.labels")= chr "asq[, 187]"
.. .. ..- attr(*, "order")= int 1
.. .. ..- attr(*, "intercept")= int 1
.. .. ..- attr(*, "response")= int 1
.. .. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. .. ..- attr(*, "predvars")= language list(asq[, 9], asq[, 187])
.. .. ..- attr(*, "dataClasses")= Named chr [1:2] "numeric" "numeric"
.. .. .. ..- attr(*, "names")= chr [1:2] "asq[, 9]" "asq[, 187]"
..- attr(*, "na.action")=Class 'omit' Named int [1:12] 26 257 352 458 508 624 820 874 1046 1082 ...
.. .. ..- attr(*, "names")= chr [1:12] "26" "257" "352" "458" ...
- attr(*, "class")= chr [1:2] "aov" "lm"
Here:
summary(test)[[1]][["Pr(>F)"]][1]
since the suggest above didn't work for me this is how i managed to solve it:
sum_test = unlist(summary(test))
then looking at the names with
names(sum_test)
i have"Pr(>F)1" and "Pr(>F)2", when the first it the requested value, so
sum_test["Pr(>F)1"]
will give the requested value
I know this is old but I looked around online and didn't find an explanation or general solution and this thread is one of the first things that comes up in a Google search.
Aniko is right, the easiest way is looking in summary(test).
tests <- summary(test)
str(tests)
That gives you a list of 1 for an independent measures aov object but it could have multiple items with repeated measures. With the repeated measures each item in the list is defined by the error term for the item in the list. Where a lot of new people get confused is that if it's between measures the one lone list item isn't named. So, they don't really notice that and don't understand why using a typical selector doesn't work.
In the independent measures case something like the following works.
tests[[1]]$'Pr(>F)'
In repeated measures it's similar but you could also use named items like...
myModelSummary$'Error: subject:A'[[1]]$'Pr(>F)'
Note there I still had to do that list selection because each one of the list items in the repeated measures model is again a list of 1.
Check out str(summary(test)) - that's where you see the p-value.
Somewhat shorter, than in BurningLeo's advice:
summary(test)[[1]][[1,"Pr(>F)"]]
summary(aov(y~factor(x)))[[1]][[5]][1]
unlist(summary(myAOV)[[2]])[[9]]
2 and 9 are the positions of p-value in myAOV model