Detecting Multiple Breaks in Financial Market Volatility Dynamics



Table 7: Testing for multiple change-points in the volatility of daily Stock Market Indices
(SMI) over the period 1989-2001

Lavielle and Moulines Test

SMI

Process

Selection Criterion

Number & Location of Breaks

FTSE

|rt|

BIC -2.616(2), -2.610(1)

LWZ -2.599(1), -2.549(0)

2

1

3/11/92, 1/8/97

1/8/97

( rt )2

BIC -2.123(1), -2.070(0)

LWZ -2.112(1), -2.069(0)

1

1

10/7/98

10/7/98

HSI

|rt|

BIC -1.121(3), -1.117(2)

LWZ -1.108(1), -1.074(0)

3

1

3/7/92, 24/1/95, 15/8/97

15/8/97

( rt )2

BIC 2.005(1), 2.009(0)

LWZ 2.010(0)

1

0

15/8/97

NIKKEI

|rt|

BIC -1.874(2), -1.867(1)

LWZ -1.857(1), -1.851(0)

2

1

15/9/92, 30/7/97

20/8/98

( rt )2

BIC -0.457(2), -0.452(1)

LWZ -0.448(0)

2

0

15/9/92, 14/10/97

S&P500

|rt|

BIC -2.525(3), -2.513(2)

LWZ -2.492(2), -2.491(1)

3

2

27/12/91, 5/1/96, 28/7/98

20/8/91, 3/2/97

( rt )2

BIC -1.602(1), -1.559(0)

LWZ -1.591(1), -1.559(0)

1

1

14/10/97

14/10/97

Notes: For brief data description refer to note 1, Table 6. The Lavielle and Moulines test is described in section 1.2. The number of
segments for multiple breaks denoted by
m is set equal to 3. The selection criteria BIC and LWZ refer to the Bayesian or Schwarz
Information Criterion and modified BIC proposed in Liu et al. (1997).

31



More intriguing information

1. The Shepherd Sinfonia
2. The name is absent
3. The name is absent
4. EXECUTIVE SUMMARY
5. The name is absent
6. The name is absent
7. Naïve Bayes vs. Decision Trees vs. Neural Networks in the Classification of Training Web Pages
8. Palvelujen vienti ja kansainvälistyminen
9. Outsourcing, Complementary Innovations and Growth
10. RETAIL SALES: DO THEY MEAN REDUCED EXPENDITURES? GERMAN GROCERY EVIDENCE