[R] ggplot2: two time series with different dates in a single plot

[Gabor Grothendieck]

On Tue, Mar 5, 2013 at 3:30 PM, Matthijs Daelman
<matthijs.daelman at gmail.com> wrote:
> Hi
>
> Using the ggplot2 package, I would like to obtain a plot that contains two
> time series that have data points on different dates.
>
> For instance, one data frame looks like:
>
> date1, value1
> 2010-01-05, 2921.74
> 2010-01-08, 2703.89
> 2010-01-14, 3594.21
> 2010-01-20, 3659.22
>
> The other data frame looks like
>
> date2, value2
> 2010-01-01, 285.85
> 2010-01-02, 229.20
> 2010-01-05, 333.91
> 2010-01-06, 338.27
> 2010-01-07, 272.85
> 2010-01-08, 249.04
> 2010-01-09, 240.07
> 2010-01-10, 255.06
> 2010-01-11, 275.42
> 2010-01-12, 252.39
>
> I would like to plot these two time series in one and the same plot, with
> date on the X axis and value on the Y axis.
>
> And while you're at it: how would you proceed to get a secondary Y axis for
> the second dataframe?
>

Lines1 <- "date1, value1
2010-01-05, 2921.74
2010-01-08, 2703.89
2010-01-14, 3594.21
2010-01-20, 3659.22"

Lines2 <- "date2, value2
2010-01-01, 285.85
2010-01-02, 229.20
2010-01-05, 333.91
2010-01-06, 338.27
2010-01-07, 272.85
2010-01-08, 249.04
2010-01-09, 240.07
2010-01-10, 255.06
2010-01-11, 275.42
2010-01-12, 252.39"

library(zoo)
library(ggplot2)

# create two zoo time series objects
z1 <- read.zoo(text = Lines1, header = TRUE, sep = ",")
z2 <- read.zoo(text = Lines2, header = TRUE, sep = ",")

# combine them into a single multivariate time series
z <- na.approx(merge(z1, z2))

# single panel
autoplot(z, facet = NULL)

# or, multiple panels with different Y axes
autoplot(z) + facet_free()

Different left and right are generally frowned upon but if you want
that anyways look at the examples in ?plot.zoo

--
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com