This paper proposes a moving average (MA) unit root test for time series data with missing observations. The proposed test statistic is asymptotically normal under the null hypothesis when applied to data with missing observations, whereas the test statistic applied to complete data is asymptotically non-normal under the null hypothesis. At the end, we provide numerical examples.

time series, MA unit root, score-type test, asymptotic normality, LBIU.