[R] Embed function strips out date index

Manussawee Sukunta msukunta at illinoisalumni.org
Wed Jun 30 22:32:42 CEST 2010


Hi,

I'm having especially hard time today and couldn't find any
clue/answer through the internet.  I hope you can help.

I'm in a process of writing a script to estimate error correction
model, and I was following an example in Bernhard Pfaff's Analysis of
Integrated and Cointegrated Time Series with R.  I have the following
price data:

> head(series,15)
           PX_SETTLE PX_SETTLE.1
2009-01-02    4515.0      925.50
2009-01-05    4540.5      927.50
2009-01-06    4603.5      930.50
2009-01-07    4470.5      905.25
2009-01-08    4474.5      906.75
2009-01-09    4430.5      885.50
2009-01-12    4402.0      868.00
2009-01-13    4343.5      868.50
2009-01-14    4130.5      839.75
2009-01-15    4070.5      839.25
2009-01-16    4129.5      848.50
2009-01-20    4032.0      806.00
2009-01-21    4018.0      836.75
2009-01-22    4011.0      825.50
2009-01-23    3998.0      823.50

Then I defined
series.d = embed(diff(series),dim=2)

which resulted in
> head(series.d,15)
        [,1]   [,2]   [,3]   [,4]
 [1,]   25.5   2.00     NA     NA
 [2,]   63.0   3.00   25.5   2.00
 [3,] -133.0 -25.25   63.0   3.00
 [4,]    4.0   1.50 -133.0 -25.25
 [5,]  -44.0 -21.25    4.0   1.50
 [6,]  -28.5 -17.50  -44.0 -21.25
 [7,]  -58.5   0.50  -28.5 -17.50
 [8,] -213.0 -28.75  -58.5   0.50
 [9,]  -60.0  -0.50 -213.0 -28.75
[10,]   59.0   9.25  -60.0  -0.50
[11,]  -97.5 -42.50   59.0   9.25
[12,]  -14.0  30.75  -97.5 -42.50
[13,]   -7.0 -11.25  -14.0  30.75
[14,]  -13.0  -2.00   -7.0 -11.25
[15,]  169.0   7.25  -13.0  -2.00

The new data series.d now has no date index.  I'm not sure how to get
it back.  I tried to xts --> order.by = index(series), but the vector
lengths are now not the same.  I feel like the answer might be
obvious, but I just can't see it.  Again, I tried searching various
forums and sites, but I couldn't find my answer.  I feel like I'm just
going around a circle.  I hope someone can help me and shed some light
on this problem.

Thank you,
Manussawee



More information about the R-help mailing list