Data cleansing algorithm for FINRA's Trade Reporting and Compliance Engine (TRACE) for python.
As of March 18, 2019 a revised sequence filter is near completion. I'm in the final stages of paramater testing which will optimize Type I and Type II error with generality in mind. My advisor and I should be submitting our a paper within the month which will condense both Nielsen's and our efforts to streamline fixed-income research, specifically for TRACE.
"Liquidity Biases in TRACE" (Dick-Nielsen 2009)
"How to clean Enhanced TRACE data" (Dick-Nielsen 2014)
"Realized Volatility, Liquidity, and Corporate Yield Spreads" (Rossi 2014)