High frequency data are often observed at irregular intervals, which complicates the analysis of lead-lag relationships between financial markets. Frequently, estimators have been used that are based on observations at regular intervals, which are adapted to the irregular observations case by ignoring some observations and imputing others. In this paper we propose an estimator that avoids imputation and uses all available transactions to calculate (cross) covariances. This creates the possibility to analyze lead-lag relationships at arbitrarily high frequencies without additional imputation bias, as long as weak identifiability conditions are satisfied. We also provide an empirical application to the lead-lag relationship between the SP500 index and futures written on it.
|Number of pages||28|
|Publication status||Published - 1995|
|Name||CentER Discussion Paper|