This paper investigates the superiority of the Yang and Zhang (YZ) estimator over the demeaned squared returns in detecting sudden breaks based on Inclan and Tiao (IT-ICSS) algorithm using Monte Carlo simulation experiments. Our findings indicate that the IT-ICSS algorithm exhibits desirable size and power properties when applied with the YZ estimator in comparison to its use with the demeaned squared returns. Empirically, we validate the superiority of the YZ estimator by relating the detected breaks with the major macroeconomic events using various US dollar exchange rates. We find that the demeaned squared returns detect many spurious breaks.