Jerrod and All,
As I see it, it’s a good thing that Marco is aware of this. He’s in a privileged position, because he’s able to see all the Reuters data.
Problem #1: We, users, are totally unaware that a problem exists. We’re unaware, because we cannot see the raw fundamentals data that comes from Reuters.
Problem #2: Let’s suppose stocks A, B, C, and D have EPS%ChgTTM values of -2, -1.25, +1.2, and NA, respectively. Then the P123 computer will assign the highest rank to stock C (because it’s +1.2), and the lowest rank to stock D (because it’s an NA)…
Solution: However, if this ranking is not what you want, then you can write “IsNA(EPS%ChgTTM,0)”. Then the P123 computer will assign the highest rank to stock C (because it’s +1.2), and the lowest rank to stock A (because it’s -2).
Problem #3: How do you add something like IsNA(EPS%ChgTTM,0)?
Solution: Marco, feel free to jump in here and tell us what the real deal is. But my guess is…
If there’s an EPS%ChgTTM > x.xx in your buy rules, then you might want to experiment, and add IsNA(EPS%ChgTTM,0) to your buy rules. Preferably just before your EPS%ChgTTM > x.xx rule.
Or, if you’ve got an EPS%ChgTTM > x.xx in your sell rules, then you might want to experiment, and add IsNA(EPS%ChgTTM,0) to your sell rules. Preferably just before your EPS%ChgTTM > x.xx rule.
Marco, feel free to jump in, and tell us what the real deal is – but my guess is, this “IsNA(xxxxxxx,0)” can be applied to all factors, all buy rules, all sell rules, and all nodes in all ranking sytems.
As I see it, the choice is up to you, the user. You, the user, don’t have to make any changes in your sims, ports, or ranking systems. However, now you’ve got a bit of additional freedom to experiment. And, who knows, you might find that, by including this IsNA(xxxxxx,0) in you buy rules, sell rules, and/or ranking systems, you may see additional improvements in the annualized returns of some or all of your ports.
I hope this helps.
Robert