Interest rate swaps and the transmission mechanism of monetary policy: A quantile connectedness approach

# INITIALISATION
nlag = 1
tau = 0.5
nfore = 20
corrected = TRUE
window.size = 200
data(cgs2021)
kable(SummaryStatistics(cgs2021))

## The following statistics are used:
## 
##           Skewness: D'Agostino, R.B. (1970). Transformation to Normality of the Null Distribution of G1. Biometrika, 57, 3, 679-681.
## 
##           Excess Kurtosis: Anscombe, F.J., Glynn, W.J. (1983) Distribution of kurtosis statistic for normal statistics. Biometrika, 70, 1, 227-234
## 
##           Normality test: Jarque, C. M., & Bera, A. K. (1980). Efficient tests for normality, homoscedasticity and serial independence of regression residuals. Economics Letters, 6(3), 255-259.
## 
##           ERS unit-root test: Elliott, G., Rothenberg, T. J., & Stock, J. H. (1996). Efficient Tests for an Autoregressive Unit Root. Econometrica, 64(4), 813-836.
## 
##           Weighted Portmanteau statistics: Fisher, T. J., & Gallagher, C. M. (2012). New weighted portmanteau statistics for time series goodness of fit testing. Journal of the American Statistical Association, 107(498), 777-787.
## 
## 

	USD1Y	EUR1Y	GBP1Y	JPY1Y
Mean	-0.001	-0.001*	-0.001***	0.000
	(0.160)	(0.074)	(0.007)	(0.805)
Variance	0.001***	0.001***	0.001***	0.000***
Skewness	-0.628***	-0.385***	-0.693***	-0.398***
	(0.000)	(0.000)	(0.000)	(0.000)
Ex.Kurtosis	13.975***	25.931***	18.756***	20.212***
	(0.000)	(0.000)	(0.000)	(0.000)
JB	31983.470***	109332.245***	57462.247***	66469.083***
	(0.000)	(0.000)	(0.000)	(0.000)
ERS	-6.358***	-23.176***	-22.380***	-21.865***
	(0.000)	(0.000)	(0.000)	(0.000)
Q(20)	90.481***	135.246***	160.961***	54.735***
	(0.000)	(0.000)	(0.000)	(0.000)
Q2(20)	1723.487***	967.321***	1522.432***	755.185***
	(0.000)	(0.000)	(0.000)	(0.000)
kendall	USD1Y	EUR1Y	GBP1Y	JPY1Y
USD1Y	1.000***	0.274***	0.280***	0.096***
EUR1Y	0.274***	1.000***	0.313***	0.085***
GBP1Y	0.280***	0.313***	1.000***	0.105***
JPY1Y	0.096***	0.085***	0.105***	1.000***

dca = ConnectednessApproach(cgs2021,
                            model="QVAR",
                            connectedness="Time",
                            nlag=nlag,
                            nfore=nfore,
                            corrected=corrected,
                            window.size=window.size,
                            VAR_config=list(QVAR=list(tau=tau)))

## Estimating model

## Computing connectedness measures

## The QVAR connectedness approach is implemented according to:
##  Chatziantoniou, I., Gabauer, D., & Stenfors, A. (2021). Interest rate swaps and the transmission mechanism of monetary policy: A quantile connectedness approach. Economics Letters, 204, 109891.

kable(dca$TABLE)

	USD1Y	EUR1Y	GBP1Y	JPY1Y	FROM
USD1Y	75.84	10.45	11.12	2.60	24.16
EUR1Y	10.21	76.47	11.14	2.18	23.53
GBP1Y	11.87	11.15	74.82	2.15	25.18
JPY1Y	5.01	2.86	2.85	89.27	10.73
TO	27.10	24.46	25.11	6.93	83.60
Inc.Own	102.93	100.94	99.93	96.20	cTCI/TCI
NET	2.93	0.94	-0.07	-3.80	27.87/20.90
NPT	2.00	3.00	1.00	0.00

PlotTCI(dca)

PlotNET(dca)

k = ncol(cgs2021)
t = nrow(dca$TCI)
date = index(cgs2021)
NAMES = colnames(cgs2021)

n = 10
quantiles = seq(0.05, 0.95, 1/n) 
print(quantiles)

[1] 0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 0.95

DCA_list = list()
TCI = array(NA, c(t, length(quantiles)), dimnames=list(as.character(tail(date,t)), quantiles))
NET = array(NA, c(t, k, length(quantiles)), dimnames=list(as.character(tail(date,t)), NAMES, quantiles))
for (j in 1:length(quantiles)) {
  dca = suppressMessages(ConnectednessApproach(cgs2021,
                              model="QVAR",
                              connectedness="Time",
                              nlag=nlag,
                              nfore=nfore,
                              corrected=corrected,
                              window.size=window.size,
                              VAR_config=list(QVAR=list(tau=quantiles[j]))))
  TCI[,j] = dca$TCI
  NET[,,j] = dca$NET
  DCA_list = c(DCA_list, list(dca$TABLE))
  print(quantiles[j])
}

[1] 0.05 [1] 0.15 [1] 0.25 [1] 0.35 [1] 0.45 [1] 0.55 [1] 0.65 [1] 0.75 [1] 0.85 [1] 0.95

nlevels = 20
threshold = 0.01 # get rid of outliers
filled.contour(tail(date,t), quantiles, TCI, xlab="", ylab="", ylim=c(0,1))

for (i in 1:k) {
  net = NET[,i,,drop=FALSE]
  net[which(net<quantile(net, threshold), arr.ind=TRUE)] = as.numeric(quantile(net, threshold))
  net[which(net>quantile(net, 1-threshold), arr.ind=TRUE)] = as.numeric(quantile(net, 1-threshold))
  
  lvls = seq(-ceiling(max(abs(net))), ceiling(max(abs(net))), length.out=nlevels)
  color.palette = function(n) {hcl.colors(n, "RdBu", rev=TRUE)}
  quantiles = as.numeric(dimnames(net)[[3]])
  filled.contour(tail(date,t), quantiles, as.matrix(net[,1,]),
                 col=color.palette(nlevels-1), levels=lvls, main=colnames(net), ylim=c(0,1))
}