Sequential Application of Factors? How?

Research
1

Hello,

I’m trying to replicate a simple 2 factor model, the wrinkle being that we want to apply a sequential application of factor ranking as follows:

A simple two factor system takes as its first (primary) factor (lets call this stage 1), the top 20% of 6 month price index history (price momentum). From the first 20% of stocks selected in stage 1, the system applies the second factor, this being to select the top 20% of price-to-book rankings (lowest).

I can’t figure out how to generate this sequential logic in P123. Everything I’m able to do using FRank is basically an intersect, ie the selected stocks are both in the top 20% of 6 month price history, and the top 20% of price-to-book rankings. Can someone throw me a bone?

Stephen

2

In the screener, you can use the #Previous scope to achieve a sequential ranking schema.

3

Ok thanks -forgot to mention in my previous post that the target of this question is the portfolio simulation. Can you tell me -does #previous work there? Thought I read somewhere that was only for screener?

4

That’s correct. #Previous will not work in a Portfolio due to the fact that buy/sell criteria are applied in parallel (i.e., to each stock at a time).

The other option is to use a whittled down universe in your Portfolio; the #Previous scope does work in universes.

5

Ok, so primary factor gets applied to the universe then do the secondary factor as a buy rule.

Thanks I’ll give that a whirl