Title: Inference on irregularly spaced time series
Authors: Fulvia Marotta - Queen Mary University of London (United Kingdom) [presenting]
Abstract: Time series data are usually recorded at regularly spaced time intervals. In spite of this, for a variety of reasons, in many fields time series can be recorded as irregularly spaced observations. Irregularly spaced time series refers to the case when the sampled time series presents irregularly spaced observations. Point estimation and large sample statistical inference are developed for time series with irregularly spaced data. Specifying the setting we make a distinction between calendar time and intrinsic, operational time. When these two times do not coincide, we have unevenly spaced elements in the sample. We first focus on the question of estimating the sample mean when data is not regularly spaced. We provide an expression for the sample mean estimator and we establish asymptotic properties and central limit theorem. Subsequently, we construct a consistent estimator for the variance of the sample mean estimator. Finite sample properties of the estimator are investigated in a Monte Carlo study which confirms the good performance of such an estimator.