Skip to main content
Financial Innovation

Table 6 MAE and MSE ratios on the real out-of-forecasts

From: Salience theory value spillovers between China’s systemically important banks: evidence from quantile connectedness

 

Return_T

Return_S

Return_C

VolatilityT

VolatilityS

Volatility_C

Skewness_T

Skewness_S

Skewness_C

Kurtosis_T

Kurtosis_S

Kurtosis_C

MAE, 1 day ahead

τ = 0.05

1.0152

1.0277

0.9931

1.0158

1.0305

1.0147

1.0021

1.0002

0.9938

1.0051

1.0084

1.0001

τ = 0.50

1.0208

1.0407

1.0014

1.0034

1.0061

0.9989

0.9942

0.9883

1.0022

0.9942

0.9926

0.9972

τ = 0.95

1.0156

1.0142

0.9987

1.0162

1.0183

0.9981

1.0095

1.0059

1.0194

1.0071

1.0006

0.9960

MAE, 22 days ahead

τ = 0.05

1.0242

1.0435

0.9655

1.0690

1.1006

0.9939

1.0195

1.0337

0.9734

1.0393

1.0740

0.9512

τ = 0.50

0.9769

0.9608

1.0022

1.0199

1.0131

1.0158

0.9720

0.9458

1.0271

0.9792

0.9652

1.0022

τ = 0.95

1.0079

1.0156

0.9906

1.0346

1.0476

1.0032

1.0245

1.0269

1.0463

1.0083

1.0002

0.9966

MSE, 1 day ahead

τ = 0.05

0.9999

1.0093

1.0397

1.0183

1.0426

1.0458

0.9813

0.9750

1.0180

0.9965

0.9983

1.0366

τ = 0.50

1.0432

1.0711

1.0018

1.0184

1.0232

0.9467

0.9610

0.9457

1.0095

1.0049

1.0004

1.0177

τ = 0.95

1.0172

1.0065

0.9820

1.0114

1.0145

1.0034

1.0051

1.0028

1.0245

0.9877

0.9815

0.9902

MSE, 22 days ahead

τ = 0.05

1.0966

1.0999

1.0319

1.1691

1.1744

1.1007

1.0832

1.0861

1.0296

1.1320

1.1390

1.0186

τ = 0.50

0.9384

0.9345

0.9897

1.0457

1.0475

0.9951

0.8811

0.8727

1.0026

0.9873

0.9853

1.0101

τ = 0.95

1.0443

1.0447

1.0033

1.0334

1.0314

1.0170

1.0192

1.0186

1.0072

1.0055

1.0017

0.9800

  1. T refers to the whole sample period; S refers to the subsample period corresponding to the Chinese stock market turbulence between January 1, 2015 and December 31, 2016; C refers to the subsample period corresponding to the COVID-19 pandemic between January 1, 2020 and June 30, 2022. Values represent ratios of HAR-RV-X model with higher-order moment interconnectedness as exogenous variables to HAR-RV-X models with STV interconnectedness as exogenous variables. A ratio above 1 suggests that the MAE/MSE of the HAR-RV-X model with STV interconnectedness as exogenous variable outperform those of the HAR-RV-X models with higher-order moment interconnectedness as exogenous variables
Back to article page