Plot multiple time series with different time indices in R using xts

Question

Using R, I have a several time series and they have different time indices, like

 date.a<-seq(as.Date('2014-01-01'),as.Date('2014-02-01'),by = 2)
 date.b<-seq(as.Date('2014-01-01'),as.Date('2014-02-15'),by = 3)
 df.a <- data.frame(time=date.a, A=sin((1:16)*pi/8))
 df.b <- data.frame(time=date.b, B=cos((1:16)*pi/8))

I merge them into one xts time series using

my.ts <- merge(xts(df.a$A,df.a$time),xts(df.b$B,df.b$time))

I want to display them both, either on separate panels, or in different colors in one.

Update:

Using xtsExtra , I was able to display both series in one plot or each in a separate panel, using the following commands:

#one panel, different colors
plot(my.ts, screens=1, type="p")
#two panels, same color
plot(my.ts, screens=c(1,2), type="p")

This approach only works with type="p" - plotting lines will not show the whole series, and create discontinuities. This is probably caused by the NA s which were caused by merging the series. Which options do I have to work with these NA s? I've found some, but I am not confident they are the best way.

1. na.omit(my.ts) and na.exclude(my.ts) seem to be too greedy - they only keep rows where both columns are not NA . How do I chance this behavior to only a rowwise operation?
2. na.approx(my.ts) fills up the NA s, but I am wondering if this linear approximation is creating artificial data points which would make the plot look differently.
3. The best way, in my opinion, would be an option which could be passed to plot to tell what should be done with NA s. Is there such a thing?

Answer 1

Use na.approx . Also note that zoo, which is already loaded by xts, is used here:

z <- na.approx(as.zoo(my.ts))

# plot.zoo
plot(z, screen = 1, col = 1:2, ylab = "Y", xlab = "")

# xyplot.zoo
library(lattice)
xyplot(z, screen = 1, col = 1:2, ylab = "Y", xlab = "")

# autoplot.zoo
library(ggplot2)
autoplot(z, facet = NULL) + ylab("Y")

Answer 2

Maybe you should do it without xts (dont see the need for it) With ggplot2 a Solution might be

    require(ggplot2)
ggplot()+geom_line(aes(x=time,y=A),data=df.a, col=2)+
  geom_line(aes(x=time,y=B),data=df.b,col=3) + ylab("")