rotating the grid to plot horizontal errors bars with Hmisc::xYplot in R

Question

I'm using xYplot to plot my regressions results with error bars. However, xYplot only plots horizontal error bars, and I need vertical error bars. Looking around for a solution, I found this thread where someone asked roughly the same question. After some messages, the user who asked the question says that "I just discovered that using the xYplot (Hmisc) and rotating the grid viewport (and the labels etc) given me exactly what I need".

So I looked around on how to rotate the grid and found that using grid library and pushviewport etc. you can rotate the grid. However, my code isn't working. Here is what I tryed so far:

estimate=structure(list(coefi = c(-5.08608456607495, -4.17906898422091, 
-2.85696514398422, -3.06968196245069, -2.73660002514793, -1.0017403629931, 
-1.66291850690335, 0.431265159072978, -0.472895611533531, 0.845421348865878, 
-0.437434919008876, 0.269041451577909, -0.233066564595661, 0.0137190330621302, 
-2.94808159763314, 1.9166875739645), lower = c(-8.1895, -6.8485, 
-5.214125, -5.532875, -5.106625, -3.271625, -3.97375, -0.09773, 
-1.340625, 0.415125, -0.86615, 0.02665125, -0.5861, -2.079, -5.626625, 
0.8115125), upper = c(-2.11475, -1.611125, -0.5602375, -0.7309625, 
-0.3721375, 1.259875, 0.7167875, 0.9672875, 0.39035, 1.30025, 
-0.05634125, 0.5115, 0.07237875, 2.14275, -0.3653, 4.202625), 
x = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 
16)), .Names = c("coefi", "lower", "upper", "x"), row.names = c("alpha.1.", 
"alpha.2.", "alpha.3.", "alpha.4.", "alpha.5.", "alpha.6.", "alpha.7.", 
"b.income", "b.democracy", "b.ginete", "b.educ", "b.patent", 
"b.fdi", "b.0", "mu.alpha", "sigma.alpha"), class = "data.frame")

legenda=c(as.character(seq(1970,2000,5)),"PIB_pc", "democ",  "legis", "educ", "patent", "FDI",  "b.0", "mu.ano", "var.ano" )

grid.newpage()
pushViewport(viewport(angle = 90, name = "VP"))
upViewport() 
xYplot(Cbind(coefi,lower, upper) ~x, data=estimate, , varwidth = TRUE, ylab="Betas",
xlab="Inclinação das Covariáveis com respectivos 95% intervalos de credibilidade \n N=409",
ylim=with(estimate,c(min(lower)-.5, max(upper)+.5)),  scales=list(cex=1.2, x = list(at=seq(1,16,     by=1), labels = legenda)) ,abline=c(list(h=0), lty="dotted", col= "grey69"), xlab.top="Adesão ao Tratado de Cooperação de Patentes, 1970-2000", draw.in = "VP")

I'd apreciate any help.

Update: there were comments pointing that the code was right. So I'm wondering if I'm not being able to communicate what I want or if it's a bug... So I'll post an imagem of the output of my code right now and you tell me if the code in you computer is giving the same output or another one:

Answer 1

You have to directly call print on lattice object (it's mentioned in grid documentation RShowDoc("grid", package="grid") - "Adding lattice to grid"):

require(grid)
grid.newpage()
pushViewport(viewport(angle = 90, name = "VP"))

print(
    xYplot(Cbind(coefi,lower, upper)~x, data=estimate, , varwidth = TRUE,
        ylab="Betas", xlab="Inclinaçao das Covariáveis com respectivos 95% intervalos de credibilidade \n N=409",
        ylim=with(estimate,c(min(lower)-.5, max(upper)+.5)),
        scales=list(cex=1.2, x = list(at=seq(1,16,by=1), labels = legenda)),
        abline=c(list(h=0), lty="dotted", col= "grey69"),
        xlab.top="Adesao ao Tratado de Cooperaçao de Patentes, 1970-2000",
        draw.in = "VP"
    ),
    newpage=FALSE
)