I uploaded an paper image. They have checked the herd behavior based on market cap.so they made the Quintile portfolio. But they restructure it every year. If i make portfolio every year then I have 5 portfolio for one year and I have 10 years data. So in total 50 portfolio are going to be emerged. But I need only 5 portfolio for the whole period. So can someone tell me what they did...is they weight them or something else. I am very much confused about this.
For each year, portfolios will remain 5 not 50. Your number of observations would be 5 too. Considering no. Of years, you need to evaluate your portfolio mix for each independent year and then compare their results. This is cross-sectional dependence to identify non-linearity, so you need to check basics of cross-sectional dependence as well.
But I need 5 portfolio for the whole period because I need to run the analysis for the whole period in one go...I want to check herd behavior. But if i restructure my portfolio every year, then in totality I have 50 portfolio. Can't I take the average of the factor which I am going to used for making portfolio. In fama- french model also, they had constructed Quintile portfolio each year for long term data but in the end they have 5 aggregate portfolio for the whole time period. How did they do that
You need to check CSAD approach and figure out methodology so you can find out how they did it.
Agree You need to check CSAD approach and figure out methodology so you can find out how they did it.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling