Hi, I am currently trying to identify the dynamic correlation between cryptocurrencies (Bitcoin and Ethereum) and financial assets (gold and S&P500). However, I am not quite sure how to interpret the results and how to get the dynamic conditional correlation graph.
mgarch dcc (d_gold_spot d_SP500 = L.d_gold_spot L.d_SP500, noconstant), arch(1) garch(1)
test _b[/Adjustment:lambda1] = _b[/Adjustment:lambda2] = 0
predict v*, variance
tsline v_d_gold_spot_d_gold_spot v_d_SP500_d_SP500
graph export "dcc_goldspot_SP500.png", replace
tsline v_d_SP500_d_gold_spot
graph export "v_d_gold_spot_d_SP500.png", replace
gen corr_goldspot_SP500 = v_d_SP500_d_gold_spot / sqrt(v_d_gold_spot_d_gold_spot * v_d_SP500_d_SP500)
tsline corr_goldspot_SP500, title("GOLD–SP500 Dynamic Conditional Correlation") ytitle("Correlation")
graph export "dcc_d_gold_spot_d_SP500.png", replace
This is the command that I used. And the graph that I got shows that gold and S&P500 don't have an obvious negative relationship, which is pretty surprising, as I expect to see a negative relationship between the two assets.
mgarch dcc (d_gold_spot d_SP500 = L.d_gold_spot L.d_SP500, noconstant), arch(1) garch(1)
test _b[/Adjustment:lambda1] = _b[/Adjustment:lambda2] = 0
predict v*, variance
tsline v_d_gold_spot_d_gold_spot v_d_SP500_d_SP500
graph export "dcc_goldspot_SP500.png", replace
tsline v_d_SP500_d_gold_spot
graph export "v_d_gold_spot_d_SP500.png", replace
gen corr_goldspot_SP500 = v_d_SP500_d_gold_spot / sqrt(v_d_gold_spot_d_gold_spot * v_d_SP500_d_SP500)
tsline corr_goldspot_SP500, title("GOLD–SP500 Dynamic Conditional Correlation") ytitle("Correlation")
graph export "dcc_d_gold_spot_d_SP500.png", replace
This is the command that I used. And the graph that I got shows that gold and S&P500 don't have an obvious negative relationship, which is pretty surprising, as I expect to see a negative relationship between the two assets.
