Re: [R] for help

2012-04-16 Thread osvald wiklander 
Hi, we have a problem concerning the use and forecasting on the GARCH (1,
1) on financial data.
 We would like to make predictions on future returns in a loop which moves
the window of estimation forward one unit for each iteration. Our code is
as follows:
 > BJORN <- read.table("C:/Users/Osvald/Desktop/DATATILLUPPSATS.txt",
+ header=TRUE, sep="\t", na.strings="NA", dec=",", strip.white=TRUE)
> t <- as.timeSeries(BJORN)
> start=length(t[2:1834])
> end=length(t)
> fore = array(0, dim=(end-start))
> for (i in start:end) {
 +fit <- garchFit(formula =garch(1,1), data = t[1:i])
 +pred <- predict(fit, n.ahead=1)
 +fore[i-start+1] = pred[,1]
+ }

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Cbind help

2012-04-16 Thread Osvald Wiklander 
Hi, we have a problem concerning the use and forecasting on the GARCH (1, 1) on 
financial data.

We would like to make predictions on future returns in a loop which moves the 
window of estimation forward one unit for each iteration. Our code is as 
follows:

> BJORN <- read.table("C:/Users/Osvald/Desktop/DATATILLUPPSATS.txt",
+ header=TRUE, sep="\t", na.strings="NA", dec=",", strip.white=TRUE)
> t <- as.timeSeries(BJORN)
> start=length(t[2:1834])
> end=length(t)
> fore = array(0, dim=(end-start))
> for (i in start:end) {
  +fit <- garchFit(formula =garch(1,1), data = t[1:i])
  +pred <- predict(fit, n.ahead=1)
  +fore[i-start+1] = pred[,1]
+ }



[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[R] Attachment

2012-04-16 Thread osvald wiklander 
Btw, in the earlier mail i forgot to mention what the error message was.
The R console tells us that: "Error in matrix(0, ncoef, ncoef) : invalid
'nrow' value (too large or NA)" before it has done any estimations.

i couldnt attach the data in the mail, so i write it manually in text.

Datum Kurs
2004-12-21 -0,001648178
2004-12-22 -0,008336043
2004-12-23 -0,010219165
2004-12-27 0
2004-12-28 0
2004-12-29 0
2004-12-30 0
2005-01-07 0,013572807
2005-01-10 -0,013572807
2005-01-11 0
2005-01-12 0
2005-01-13 -0,005201194
2005-01-14 0,005201194
2005-01-17 0
2005-01-18 0
2005-01-19 -0,005201194
2005-01-20 0
2005-01-21 -0,008809521
2005-01-24 -0,008991925
2005-01-25 -0,005485898
2005-01-26 -0,014977156
2005-01-27 0
2005-01-28 0,007553138
2005-01-31 0
2005-02-01 0
2005-02-02 0,028963696
2005-02-03 0
2005-02-04 0
2005-02-07 0,00694886
2005-02-08 -0,00694886
2005-02-09 0,00694886
2005-02-10 -0,003460532
2005-02-11 -0,003488328
2005-02-14 0,00694886
2005-02-15 0,016900751
2005-02-16 0,003302636
2005-02-17 -0,003302636
2005-02-18 0
2005-02-21 0,05299042
2005-02-22 0,022862883
2005-02-23 -0,001394205
2005-02-24 0,013744643
2005-02-25 -0,001355054
2005-02-28 0,006733383
2005-03-01 0,001334239
2005-03-02 -0,006712568
2005-03-03 0,004039978
2005-03-04 -0,005395032
2005-03-07 -0,002722858
2005-03-08 -0,005497472
2005-03-09 0,002757435
2005-03-10 -0,005532489
2005-03-11 -0,005603878
2005-03-14 0
2005-03-15 0,001407763
2005-03-16 -0,015742215
2005-03-17 0,025470819
2005-03-18 0,026652195
2005-03-21 0,017728767
2005-03-22 0,014640526
2005-03-23 -0,00240607
2005-03-24 -0,004852503
2005-03-29 -0,004907334
2005-03-30 -0,015063746
2005-03-31 0,010100327
2005-04-01 0,008649109
2005-04-04 -0,006160309
2005-04-05 -0,010041938
2005-04-06 -0,002547189
2005-04-07 0
2005-04-08 0,015063746
2005-04-11 -0,00496342
2005-04-12 0
2005-04-13 -0,005020801
2005-04-14 0,009984221
2005-04-15 -0,009984221
2005-04-18 -0,014114209
2005-04-19 0,003895044
2005-04-20 0
2005-04-21 -0,005201194
2005-04-22 0,002608383
2005-04-25 0,00773245
2005-04-26 0,015063746
2005-04-27 -0,017625963
2005-04-28 -0,013042857
2005-04-29 0
2005-05-02 -0,005328834
2005-05-03 0
2005-05-04 0,005328834
2005-05-06 0,013042857
2005-05-09 -0,013042857
2005-05-10 0
2005-05-11 0
2005-05-12 -0,002656244
2005-05-13 0,023340843
2005-05-16 0,009984221
2005-05-17 -0,003717237
2005-05-18 0,004949278
2005-05-19 0,04105702
2005-05-20 0,02606549
2005-05-23 -0,002113359
2005-05-24 -0,010723865
2005-05-25 -0,010995384
2005-05-26 0,004431615
2005-05-27 0,010885143
2005-05-30 -0,004321374
2005-05-31 0,002166062
2005-06-01 -0,002166062
2005-06-02 0,008600172
2005-06-03 -0,012964977
2005-06-07 0,021398145
2005-06-08 -0,025807264
2005-06-09 -0,00222146
2005-06-10 0,015316758
2005-06-13 -0,004321374
2005-06-14 -0,00656377
2005-06-15 -0,008908917
2005-06-16 0,006698762
2005-06-17 0,007686829
2005-06-21 -0,007686829
2005-06-22 0
2005-06-23 0,004409119
2005-06-27 -0,011107881
2005-06-28 0
2005-06-29 0,008908917
2005-06-30 -0,006664496
2005-07-01 0,004454341
2005-07-04 0
2005-07-05 -0,004454341
2005-07-06 -0,004500501
2005-07-07 0
2005-07-08 0,019894829
2005-07-11 0,002155312
2005-07-12 0
2005-07-13 0,004278798
2005-07-14 0,004237053
2005-07-15 0
2005-07-18 0,006279066
2005-07-19 0,002073009
2005-07-20 0,012234456
2005-07-21 -0,00811789
2005-07-25 0,00811789
2005-07-26 0,021572106
2005-07-27 0,005701979
2005-07-28 -0,011479818
2005-07-29 0
2005-08-01 0
2005-08-02 -0,007825338
2005-08-03 0,019305155
2005-08-04 0
2005-08-05 0
2005-08-08 -0,003792989
2005-08-09 0,003792989
2005-08-10 0,027438248
2005-08-11 0,032437843
2005-08-12 0
2005-08-15 0,008148353
2005-08-16 0,001611484
2005-08-17 0
2005-08-19 0,029534079
2005-08-22 -0,013739811
2005-08-23 0,027058233
2005-08-24 -0,027058233
2005-08-25 0,015239967
2005-08-29 -0,009079658
2005-08-30 -0,006160309
2005-08-31 0,006160309
2005-09-01 -0,003069232
2005-09-02 -0,015680204
2005-09-05 0,00476379
2005-09-06 0,012453742
2005-09-07 0,004579598
2005-09-08 -0,009208002
2005-09-09 -0,006248949
2005-09-12 0,018483406
2005-09-13 -0,012234456
2005-09-14 0
2005-09-15 0
2005-09-16 0,003091077
2005-09-19 0,003069232
2005-09-20 -0,006160309
2005-09-21 0,003091077
2005-09-22 0
2005-09-23 0,003069232
2005-09-26 -0,010838571
2005-09-27 0,010838571
2005-09-28 0,003047693
2005-09-29 0,003026455
2005-09-30 0,00300551
2005-10-03 0,008893713
2005-10-04 0,014429715
2005-10-05 -0,002847843
2005-10-06 -0,014526253
2005-10-07 0,005868934
2005-10-10 0,007226364
2005-10-11 -0,010150917
2005-10-12 0,002924553
2005-10-13 -0,005868934
2005-10-14 -0,01350247
2005-10-17 0,007553138
2005-10-18 -0,010611562
2005-10-19 -0,017217531
2005-10-20 0,018749436
2005-10-21 -0,006160309
2005-10-25 0,006160309
2005-10-26 0
2005-10-27 0,009079658
2005-10-28 0,014723257
2005-10-31 -0,023802915
2005-11-01 0,009079658
2005-11-02 0,011818266
2005-11-03 0,00865732
2005-11-04 -0,005752329
2005-11-07 0,011429462
2005-11-08 0,008378932
2005-11-09 0,034549269
2005-11-10 -0,00773245
2005-11-11 0,002592811
2005-11-14 -0

[R] Dividing tick-data into intervalls

2012-05-08 Thread osvald wiklander 




Hi everybody, I am sorry that I am kind of spamming this forum, but I have 
searched for some input everywhere and cant really find a nice solution for my 
problem.

Data looks like:

   price
2011-11-01 08:00:00  0.0
2011-11-01 08:00:00  0.0
2011-11-01 08:02:00  0.0
2011-11-01 08:03:00 -0.01709
2011-11-01 08:13:00  0.000690001
2011-11-01 08:24:00  0.000658411
2011-11-01 08:29:00  0.0
2011-11-01 08:29:00  0.0
2011-11-01 08:29:00  0.0
2011-11-01 08:29:00  0.0
2011-11-01 08:29:00  0.002166062
2011-11-01 08:44:00  0.0
2011-11-01 08:44:00  -0.002166062
2011-11-01 08:44:00  0.004321374
2011-11-01 10:36:00  0.010618976
2011-11-01 15:59:00  0.002092990

So the price column in fact shows the difference in the price of the equity. It 
is about 100 days. 

I want to be able to convert this data to regurly spaced intervals, like every 
fifth minute the previous tick is returned.

So, for example I want that the new time serie should look like:

   price
2011-11-01 08:00:00  0.0
2011-11-01 08:05:00 -0.01709
2011-11-01 08:10:00 -0.01709
2011-11-01 08:15:00  0.000690001
2011-11-01 08:20:00  0.0
2011-11-01 08:25:00  0.000658411
2011-11-01 08:30:00  0.002166062

And so on for the 100 days. And the I want to do this for ten minutes, 30 
minutes etc.

Earlier I have tried with aggregatePrice() (in the RTAQ package) combined with 
a for() loop. However this didnt work so well. 

I assume it exist some easy way to do this? However, as I said, I have searched 
for a solution without any luck.

So, what approach should I have? Have any of you done something like this 
before?

Thanks in advance Reply With 
Quote
[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.