Inspired by https://www.quantopian.com/posts/quantpedia-trading-strategy-series-reversals-during-earnings-announcements - I also thought to try to dive into another paper.
I would like to understand another paper that I have looked at - with http://www.bengrahaminvesting.ca/Research/Papers/When_Two_Anomalies_Meet.pdf
by Zhipeng Yan & Yan Zhao
A trading strategy of taking a long position in value stocks when both EARs and earnings surprises are positive and a short position in glamour stocks when both are negative can generate 16.6% to18.8% annual returns.
Main findings
With our simulation where data overlap, we have evidence to show strong showing of the strategy, supporting a similar level of spread. However, when extending time period to more current time period, we often find that the advantage drops dramatically since 2009 (when draft of paper was published)
Asking for help
Would love to get some comments and suggestion on the research. I have used various approximation on the fundamental data that is provided by morning-star to approximate the data from the research, would like to know better suggestion on how to reproduce the data
Also, any ideas if there's still different way we could squeeze out something in recent years? I'm surprise that spread was squeezed out around 2009 so quickly after the paper is published (or maybe I shouldnt be)
Any other similar researches that we can apply the same framework for studying?
Thanks all!