Stratified sampling vs optimization

[cfacandid]

Comparing to full replication, stratified sampling has obvious advantages: lowest costs, possibility of indexing to large component containing index.
As for optimization it is more costly than stratified sampling and tracking error can be higher as risk factors are dynamic and can be wrongly modelled and so on.
So I know about optimization drawbacks, but what are its advantages? Are there any? I did not find in the texts.

 

[kurmanal]

Quoting the text: “An advantage of optimization compared with stratified sampling is that optimization takes into account the covariances among the factors used to explain the return on stocks. The stratified sampling approach implicitly assumes the factors are mutually uncorrelated.”
(Institute 199)
Institute, CFA. Level III 2013 Volume 4 Fixed Income and Equity Portfolio Management. John Wiley & Sons P&T, 6/18/2012. <vbk:9781937537364#page(199)>.

 

[nvestn]

It has been stated that optomization has lower tracking error than stratified sampling.

 

[cfacandid]

Thank you. You help a lot.
As a matter of fact they also write optimization uderstates expected tracking risk and after the fact the effect are similar to stratified sampling (volume 4 page 199).
They also write that optimization is an alternative to full replication if you have separate accounts of smaller sizes (not enough cash to buy all stock to replicate).