Skip to main content

Table 31 Selection of the best parametric model as a function of s, M and \(M'\) for the measure \(R_t^{s}\) computed on Netflix simulated data

From: Drawdown-based risk indicators for high-frequency financial volumes

Model selection for \(R_t^{s}\)-simulated data for NETFLIX

M=80%

Lognormal

 

Exponential

 

Weibull

 

Gamma

 

\(M' \mid s=0\)

AIC

BIC

AIC

BIC

AIC

BIC

AIC

BIC

\(30\%\)

1.1309e+03

1.1366e+03

1.3318e+03

1.3346e+03

1.1715e+03

1.1772e+03

1.2732e+03

1.2789e+03

\(40\%\)

1.0610e+03

1.0667e+03

1.2630e+03

1.2959e+03

1.1055e+03

1.1112e+03

1.1870e+03

1.1926e+03

\(50\%\)

978.2024

983.8749

1.1597e+03

1.1626e+03

1.0236e+03

1.0293e+03

1.0825e+03

1.0882e+03

\(M' \mid s=5\)

AIC

BIC

AIC

BIC

AIC

BIC

AIC

BIC

\(30\%\)

1.0959e+03

1.1016e+03

1.3214e+03

1.3242e+03

1.1465e+03

1.1522e+03

1.2793e+03

1.2850e+03

\(40\%\)

1.0309e+03

1.0366e+03

1.2723e+03

1.2751e+03

1.0858e+03

1.0915e+03

1.1989e+03

1.2046e+03

\(50\%\)

1.0059e+03

1.0115e+03

1.2075e+03

1.2104e+03

1.0550e+03

1.0607e+03

1.1276e+03

1.1332e+03

\(M' \mid s=50\)

AIC

BIC

AIC

BIC

AIC

BIC

AIC

BIC

\(30\%\)

1.0231e+03

1.0287+03

1.1597e+03

1.1625e+03

1.0626e+03

1.0683e+03

1.1151e+03

1.1208e+03

\(40\%\)

982.4874

988.1599

1.1393e+03

1.1421e+03

1.0278e+03

1.0334e+03

1.0854e+03

1.0911e+03

\(50\%\)

943.4720

949.1446

1.0674e+03

1.0703e+03

983.4023

989.0749

1.0241e+03

1.0298e+03

\(M' \mid s=100\)

AIC

BIC

AIC

BIC

AIC

BIC

AIC

BIC

\(30\%\)

902.1937

907.8862

1.0443e+03

1.0472e+03

948.8200

954.4925

993.3576

999.0302

\(40\%\)

860.9473

866.6199

1.0128e+03

1.0157e+03

913.0693

918.7419

961.4471

967.1197

\(50\%\)

817.5565

823.2290

895.3954

898.2317

855.7938

861.4664

873.7375

879.4100

  1. The best parametric model is chosen by means of the AIC and BIC criteria. The smallest AIC and BIC values are in bold