| Project/Area Number |
13630029
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
MORIMUNE Kimio Kyoto University, Graduate School of Economics, Professor, 経済学研究科, 教授 (20109078)
|
|---|
| Project Period (FY) |
2001 – 2004
|
| Project Status |
Completed (Fiscal Year 2004)
|
| Budget Amount *help |
¥3,900,000 (Direct Cost: ¥3,900,000)
Fiscal Year 2004: ¥700,000 (Direct Cost: ¥700,000)
Fiscal Year 2003: ¥700,000 (Direct Cost: ¥700,000)
Fiscal Year 2002: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2001: ¥1,600,000 (Direct Cost: ¥1,600,000)
|
|---|
| Keywords | autoregressive moving average / unit root / cointegation / volatility / run test / ベクトル時系列 / Bootstrap / Nonparametric / ブラウン運動 / 長期均衡 / 構造変化 / トレンド |
|---|
| Research Abstract |
Dickey and Fuller proposed the tests for unit root hypotheses in a uni-variate time series. Perron(1989) extended the t-ratio type unit-root tests so that they allow for a break in the deterministic trend and/or in the intercept term. In practice, it seems difficult to specify the break point correctly. Zivot and Andrews (1992) proposed a test in which the break point is statistically determined but their test does not necessarily lead to an empirically satisfactory break point. Morimune and Nakagawa (1999) studied the effect of a misspecified break point on the Perron tests, and the accuracy of the asymptotic expression is examined under various specifications of the error. This paper proposes to set a break interval that possibly covers a break point in the Perron tests. It is difficult to specify a break point but easier to set a break interval in empirical studies. Tests lose power by setting break intervals, but it helps to avoid misspecifying the break point. The unit root test is less susceptible to the choice of a particular break point. Test is applied to the US macro series. In this application, the break intervals used in our study is varied from the shortest 1930-1930 to the longest 1930-1941 interval in most cases.
|
