[R] [r] How to obtain the correlation coefficients between two variables in the R object
Michael Dewey
||@t@ @end|ng |rom dewey@myzen@co@uk
Mon Nov 25 14:40:57 CET 2019
To get multiple plots on the same physical page try
?layout
ot look through the various option under
?par
I do not know enough about DCC models to answer how you get correlations.
Michael
On 25/11/2019 10:34, Subhamitra Patra wrote:
> Dear R-users,
>
> I am estimating the DCC-model by using the following code and successfully
> making the individual plots indicating the dynamic correlation between the
> two variables stored in the R data frame (namely dcc.fit in my code). My
> code is
>
> Dat = Date2[,2:13, drop = FALSE]
> library(rmgarch)
> library(rugarch)
> xspec = ugarchspec(mean.model = list(armaOrder = c(1, 1)), variance.model =
> list(garchOrder = c(1,1), model = 'eGARCH'), distribution.model = 'norm')
> uspec = multispec(replicate(12, xspec)) # Define a function using xspec
> dcc.garch11spec = dccspec(uspec = uspec, dccOrder = c(1, 1), distribution =
> 'mvnorm')
> dcc.fit = dccfit(dcc.garch11spec, data = Dat)
> r1=rcor(dcc.fit, type="R")
> print(r1)
>
> Here, I need to obtain the DCC relationship between the two variables. I
> successfully make a separate plot for the correlation between the two
> variables with the following code.
> plot(rcor(dcc.fit, type="R")['A1','A2',], type='l')
> plot(rcor(dcc.fit, type="R")['B1','B2',], type='l')
> plot(rcor(dcc.fit, type="R")['C1','C2',], type='l')
> plot(rcor(dcc.fit, type="R")['D1','D2',], type='l')
> plot(rcor(dcc.fit, type="R")['E1','E2',], type='l')
> plot(rcor(dcc.fit, type="R")['F1','F2',], type='l')
>
> Here, I request your help on two things which I am unable to do.
>
> 1. I need to arrange all the above individual plots (counted as 6) on one
> page.
> 2. I need to obtain the correlation coefficients between the particular two
> variables (i.e. A1 versus A2, B1 versus B2, .......so on) throughout the
> entire period.
>
> For your convenience, I have attached my sample data.
>
> I definitely expect a positive help from the experts as the suggestions
> from the expert through this R forum has always solved my previous problem.
>
> Thank you.
>
--
Michael
http://www.dewey.myzen.co.uk/home.html
More information about the R-help
mailing list