Journal of Applied Economic Sciences
Volume IV/ Issue 1(7)/ Spring 2009
Table 1. Number of observations, frequency of crises and tranquil periods in both stock markets and currency
markets categorized by country
|
Country |
Currency Market |
Stock Market | |||||
|
Obs. |
Crises |
Tranquil |
% Crises |
Crises |
Tranquil |
% Crises | |
|
Argentina |
39 |
3 |
36 |
8.33% |
0 |
39 |
0.00% |
|
Brazil |
83 |
0 |
83 |
0.00% |
2 |
81 |
2.47% |
|
Bulgaria |
95 |
0 |
95 |
0.00% |
8 |
87 |
9.20% |
|
Chile |
68 |
0 |
68 |
0.00% |
0 |
68 |
0.00% |
|
China |
68 |
3 |
65 |
4.62% |
6 |
62 |
9.68% |
|
Colombia |
68 |
2 |
66 |
3.03% |
0 |
68 |
0.00% |
|
Croatia |
95 |
1 |
94 |
1.06% |
2 |
93 |
2.15% |
|
Hungary |
78 |
0 |
78 |
0.00% |
0 |
78 |
0.00% |
|
India |
43 |
0 |
43 |
0.00% |
1 |
42 |
2.38% |
|
Indonesia |
43 |
1 |
42 |
2.38% |
0 |
43 |
0.00% |
|
Israel |
49 |
0 |
49 |
0.00% |
0 |
49 |
0.00% |
|
Malaysia |
82 |
2 |
80 |
2.50% |
0 |
82 |
0.00% |
|
Mexico |
83 |
2 |
81 |
2.47% |
0 |
83 |
0.00% |
|
Peru |
59 |
1 |
58 |
1.72% |
2 |
57 |
3.51% |
|
Philippines |
77 |
1 |
76 |
1.32% |
2 |
75 |
2.67% |
|
Poland |
95 |
3 |
92 |
3.26% |
4 |
91 |
4.40% |
|
Russia |
95 |
2 |
93 |
2.15% |
1 |
94 |
1.06% |
|
South Africa |
95 |
3 |
92 |
3.26% |
0 |
95 |
0.00% |
|
South Korea |
95 |
0 |
95 |
0.00% |
1 |
94 |
1.06% |
|
Thailand |
77 |
0 |
77 |
0.00% |
0 |
77 |
0.00% |
|
Turkey__________ |
77 |
0 |
77 |
0.00% |
9 |
68 |
13.24% |
|
Total |
1564 |
24 |
1540 |
1.56% |
38 |
1526 |
2.49% |
4.2. Stock Market Crisis Definition
To identify the crises periods in stock markets, we follow Mishkin and White (2002) definition
of stock market crisis as falls in price of an index below some threshold during a specified period of
time. We take the threshold to be 25 percent and the period to be 6 months. Country’s main stock
market indexes have been used as indicators of the stock market situation. Table 2 in Appendix shows
the stock market indices used for respective countries. We use the following formula to identify crises
in the emerging stock markets:
crisisi,t
'1, if (P,t/P-6)-1 <=-0.25
(5)
0, otherwise
< 7
where crisisi,t is the stock market crisis indicator of an emerging country denoted by i at time t, Pi,t
is the price of the index at time t while Pi.t-6 is the price of the index six months ago. The formula
considers a crisis to be a drop in the index of more than a quarter of its value six month ago. Using
the above definition, we identify a total number of 38 crises in our sample data. Table 1 above shows
frequencies of crises and tranquil periods in respective countries.
4.3. Model Specification
Three models are tested to see whether changes in sovereign credit default swaps improve
forecasts in emerging stock markets and currency crises respectively. The base model includes
variables usually used in literature of predicting crises in emerging markets. The second model adds
changes in sovereign CDS premiums to the base model, while the third tests the significance of a
model which includes only changes in CDS premiums as explanatory variable. Variables in all three
models are lagged by one month to check the predictability a crisis a month in advance. In all the
regression models for both markets, we use a logistic function of the type:
129
More intriguing information
1. The name is absent2. Solidaristic Wage Bargaining
3. The name is absent
4. The name is absent
5. Insurance within the firm
6. The name is absent
7. The name is absent
8. Natural hazard mitigation in Southern California
9. Types of Tax Concessions for Promoting Investment in Free Economic and Trade Areas
10. The name is absent