Skip to content

Converting to weekly timestamps removes years from chart.TimeSeries (and its extensions)

4 messages · Brian G. Peterson, Joshua Ulrich, Ilya Kipnis

#
So something that just irritated me is this:

I'm replicating a strategy that uses weekly rebalancing, but when I go to
plot the equity curve, the years cut off on the x axis, which looks weird.

Here's my code:

require(quantstrat)

#long TLT 60%, leverage 3x to 180%, rebalance weekly, short VIX 40%
getSymbols("^VIX", from="1990-01-01")
getSymbols("TLT", from="1990-01-01")
VIX <- to.weekly(OHLC(VIX), OHLC=TRUE)
TLT <- to.weekly(OHLC(TLT), OHLC=TRUE)
vixRets <- Cl(VIX)/Op(VIX)-1
tltRets <- Cl(TLT)/Op(TLT)-1
both <- merge(vixRets, tltRets, join='inner')
colnames(both) <- c("VIX", "TLT")
portfRets <- -.4*both$VIX+1.8*both$TLT
colnames(portfRets) <- "VIXTLT"
charts.PerformanceSummary(portfRets)

Along with the attached picture.

Any tips on how to get the years back into the x axis would be appreciated.

Thanks.

-Ilya
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20140928/fb364775/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: StructArb.png
Type: image/png
Size: 29654 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20140928/fb364775/attachment.png>
#
maybe RTFM?

if you follow the seeAlso links in the pdf or online help, you'll find

?chart.TimeSeries

see

data.format


charts.PerformanceSummary(portfRets,date.format="%m/%y")
On 09/28/2014 08:40 PM, Ilya Kipnis wrote:
#
I get a very reasonable chart and axis when I run your code in a clean
session.  The years aren't simply cut off the x-axis labels, all the
"months" are "Aug" and the grid lines extend beyond the edges of the
plots.  I think something is funky with your R session.
--
Joshua Ulrich  |  about.me/joshuaulrich
FOSS Trading  |  www.fosstrading.com
On Sun, Sep 28, 2014 at 8:40 PM, Ilya Kipnis <ilya.kipnis at gmail.com> wrote:
#
Well, this is really irritating. I just loaded up dplyr and suddenly got
those same issues, but when I didn't have it loaded, everything went
smoothly.

Very strange.

On Sun, Sep 28, 2014 at 10:22 PM, Joshua Ulrich <josh.m.ulrich at gmail.com>
wrote: