White paper — Portfolio Optimization: Thinking ‘Outside the Style Box’

(full ver­sion)

Down­load White Paper

Swan Research — Innovative Portfolio Optimization

Swan is focused on help­ing provide finan­cial advi­sors with the thought lead­er­ship nec­es­sary to dif­fer­en­ti­ate them­selves and make their busi­ness­es stronger and more valu­able. The pur­pose of this doc­u­ment is to high­light our the­o­ret­i­cal view that a diver­si­fied hedged assets port­fo­lio is a more effec­tive and effi­cient way to opti­mize a port­fo­lio than tra­di­tion­al port­fo­lio opti­miza­tion.

The goal will be to present evi­dence to sup­port the fol­low­ing port­fo­lio man­age­ment per­spec­tives:
• Tra­di­tion­al port­fo­lio opti­miza­tion is flawed and poten­tial­ly mis­lead­ing and the effi­cient fron­tier is of lim­it­ed use
• Tra­di­tion­al port­fo­lio opti­miza­tion leads to fair­ly indis­tin­guish­able asset allo­ca­tions
• Tra­di­tion­al port­fo­lio opti­miza­tion fails to min­i­mize loss­es, as they are built to min­i­mize volatil­i­ty
• An alter­na­tive approach to port­fo­lio opti­miza­tion (such as the Defined Risk Strat­e­gy), that direct­ly address­es
mar­ket risk, can lead to more effec­tive and effi­cient port­fo­lios
• Port­fo­lio results can be improved through the use of hedged assets
• A defined risk port­fo­lio, built upon the con­cept of max­i­miz­ing return while min­i­miz­ing an investors’ pos­si­ble
lev­el of “pain”, could intro­duce a par­a­digm shift away from tra­di­tion­al port­fo­lio opti­miza­tion

Mean Vari­ance Opti­miza­tion (MVO) is cur­rent­ly the most com­mon method­ol­o­gy for cre­at­ing port­fo­lios based on MPT. In essence, the process math­e­mat­i­cal­ly deter­mi­nes the opti­mal weight­ings amongst a list of avail­able asset class­es to yield port­fo­lios that have the high­est expect­ed
return for a given lev­el of risk. This is com­put­ed based on the inputs of the returns and stan­dard devi­a­tions of each of the avail­able assets, and the cor­re­la­tions amongst those assets. While MVO is a pop­u­lar tool for cre­at­ing effi­cient port­fo­lios, it can be a vic­tim of its own inef­fi­cien­cies.”  (The Kit­ses Report, July 2008, Michael Kit­ses)

Port­fo­lios cre­at­ed via MVO favor indi­vid­u­al assets with high return-to-risk esti­mates. There­fore, using MVO tends to mag­ni­fy errors in esti­mates. Port­fo­lio opti­miza­tion is built upon the assump­tion that asset class­es will con­tin­ue to exhibit past pat­terns of return, cor­re­la­tion, and vari­ance.

How­ev­er, as expe­ri­enced in 2008 and seen in Exhibit 1, past pat­terns do not always per­sist. This can lead to out­comes very dif­fer­ent than the expec­ta­tions.  Note the long-term cor­re­la­tion of assets pri­or to the Finan­cial Cri­sis (cor­re­la­tion mea­sures how close­ly two dif­fer­ent invest­ments move in con­junc­tion with one anoth­er; if one is seek­ing to diver­si­fy an invest­ment port­fo­lio, low­er cor­re­la­tions or neg­a­tive cor­re­la­tions are desired).

Swan believes it is impor­tant that investors have a bet­ter under­stand­ing of Mod­ern Port­fo­lio The­o­ry, how it is broad­ly used to con­struct port­fo­lios, and the impli­ca­tions of apply­ing MPT ver­sus dif­fer­ent approach­es to port­fo­lio con­struc­tion when con­sid­er­ing their own invest­ments, port­fo­lios and long-term goals.

This paper endeav­ors to provide a broad­er under­stand­ing and the alter­na­tive approach to port­fo­lio con­struc­tion and opti­miza­tion inher­ent in the Defined Risk Strat­e­gy.