Enhanced indexing generates alpha and control/minimize track risk and has highest information ratio. Where the indexing with optimization has lower tracking risk than stratified random sampling but might not generate alpha.So the tracking risk could still be higher than enhanced indexing.
If so, does indexing with optimization has any advantage over enhanced indexing?
Thank you!
I think there are 3 approaches:
Full indexing (passive)
Enhanced indexing (semi-active)
Full Active management (not indexing)
The first has zero alpha and zero tracking error and may be costly due to rebalancing especially for large indices. The second has highest IR, it is less costly but strategy may be followed from other managers until alpha disappear. The third may have highest alpha but with rapidly increased active risk and high alpha strategy might not be sustainable in the long run.
I’m not 100 % sure but I suppose that optimization and stratified sampling are both part of enhanced indexing strategy.
Stratified Sampling and Optimization are methods to reduce the burdens of full index replication (large number of and/or illiquid index constituents for example). As such I believe they are more derivatives of pure indexing than enhanced indexing. Enhanced indexing (when looking at equities) is a moderation of active management where a manager is allocated close to an index but overweights and underweights positions based on views of expected returns. Where this gets confusing is enhanced indexing is also covered in Fixed Income, where it can have more of a stratified sampling form (matching primary risk factors) as well as a form of enhanced indexing similar to the equity section definition (small risk factor mismatches).
I’m still not totally clear on where all these worlds collide, but there is some distinction between the Fixed Income and Equity readings that I think is a good idea to follow and keep separate in your head.
Makes sense if we look at optimization and stratified sampling under enhanced indexing. Of course, there is FI as well… Thank you both for the help like always!
I brought thi up because of EOC question listing both enhanced indexing and indexing using optimization as options (Question 18 of Equity). That is why I was not quite sure the classification of indexing using optimization.
in terms of equity mgt, i think optimization and stratified sampling fell more under passive than semi-active, at least this is how it is structured in the texts although they seems like part of semi-active strategy.
not sure…
Yes, certainly. Again, from the equity perspective (and probably fixed income just approached somewhat differently in the text), the goal of both of these is to overcome obstacles that can be present in full index replication, which is a passive objective.
Think of the portfolio manager who wants to build a portfolio that replicates an index that contains 5,000 securities, roughly 10% of which have low liquidity. Sampling and Optimization can be employed to overcome the problems of a large number of holdings and imperfect investability (liquidity), but both will introduce tracking error. This isn’t tracking error due to active decisions, but tracking error due simply to not perfectly matching the index.
Contrast that to enhanced indexing, where the objective is to earn some alpha by tilting holding weights one way or the other vs the index based on the manager’s views. Now we’re getting into the realm of semi-active and away from passive. There is tracking risk here too, but now it’s active tracking risk, as it’s risk related to active decisions.
Well explained. Bravo!
Great summary! Thumbs up!
very good point!
great explaination!