search

vickyyin1493 / Replication-FFJR

0 stars 0 forks source link

Problem 3: The plot we replicated for table 3 is a bit weird. #8

Closed iangow closed 1 year ago

iangow commented 1 year ago

@vickyyin1493 @shizhe75 What are you expecting to see with this plot? Are you trying to replicate a plot from FFJR?

vickyyin1493 commented 1 year ago

We were trying to use it to support (or not) theory A' from the tutorial. As we were thinking by replicating using finer dates we were really trying to distinguish between the theories?

We were expecting to see a positive relationship between lagged returns and splits. As we thought this would mean splits are usually followed by a period of higher returns.

The codes we got were from tutorial solutions.

@iangow @shizhe75

iangow commented 1 year ago

I see now. But the plot in the solutions (this comes from the paper in some form) is aggregating over the whole market by year. Aggregating by day is not likely to do much (most days will have either no splits or one split) and the plot you see makes perfect sense given this.

I think the later figure would be a better one to use.