How to get correlation using mgarch dcc

[Morimoto Susumu]

Several papers use Dynamic conditional correlation multivariate GARCH models to get conditional correlation. One example is Economics Letters 91 (2006) 110–116, The comovement between output and prices: Evidence from a dynamic conditional correlation GARCH model written by Jim Lee. Is there anybody who knows how to get the correlation using mgarch dcc?
Please help me out!

[Attaullah Shah]

May be I am late to the party, but someone might still be interested in knowing the answer. After you estimate the mgarch equation, which is often of the following structure,

mgarch dcc (variable1 variable2=), arch(1) iterate(1000)

then you can predict the conditional variance and covariance with the following command;

predict h2* if e(sample), variance

This produces three variables,

1. h_variable1_variable1 , which is the conditional variance of the variabel1
2. h_variable2_variable2, which is the conditional variance of the variable 2
3. h_variaabe1_variable2, which is the conditional covariance between variable and variable

[beyza oktay]

when i do all these written steps and get the correlation, i tsline the correlation to see the graph but actually it is not the same as usual correlation graphs on the papers. Instead it is like the return graphs... I dont know what is wrong

[anduong]

Please provide your code. Please note that you need to calculate the correlation yourself. The code "predict h2* if e(sample), variance" provides only the variance-covariance, not the correlation.