Small Sample Properties Of Unit Root Tests For A Random Walk Model With AR (1) Disturbances

スポンサーリンク

概要

• 論文の詳細を見る

• We conduct a small Monte Carlo simulation to find empirical percentiles and powers of augmented Dickey-Fuller tests and Phillips-Perron tests for ARIMA(1,1,0) null hypothesis against ARIMA(1,0,0)×(1,0,0) alternatives. We derive Phillips-Perron tests using parametric estimation of long run variance. In particular, we focus our analysis on the effect of initial observations and sample sizes. We find that Phillips-Perron tests are slightly more powerful than augmented Dickey-Fuller tests in general and non-zero initial observation works to strengthen the powers of the tests even in small sample sizes like 20 or 50.

• 関西学院大学の論文

著者

• Fujiki Kiyoshi

  Graduate School Of Business Administration Kwansei Gakuin University

• Sugihara Soichi

  School of Business Administration, Kwansei Gakuin University

• Sugihara Soichi

  School Of Business Administration Kwansei Gakuin University

関連論文

• Small Sample Properties Of Unit Root Tests For A Random Walk Model With AR (1) Disturbances

スポンサーリンク