Full Markowitz portfolio selection requires millions of pairwise correlation estimates.

Evidence

Markowitz agreed to become the informal supervisor of Sharpe’s UCLA Ph.D. thesis. Despite Markowitz’s sense that his work was a “closed logical piece,” there was one issue that he had broached but not fully pursued. His full version of portfolio selection required getting securities analysts to estimate the correlation of every pair of securities among which selection was to be made. The number of such correlations increased rapidly with the number of securities being analyzed. Selection among 1,000 securities, for example, would require estimation of 499,500 correlations (Baumol 1966, pp. 98-99). — Donald MacKenzie, An Engine, Not a Camera, 229

Categorization

• CANDY: The 499,500 correlations for 1,000 securities is a nice dinner-party numerical fact that makes the computational hurdle visceral.
• XREF: Connects to the broader performativity literature — computational tractability as a constraint on which models can become market practice. — connects to: Computational and data limits, not theory, made early portfolio selection unusable. — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 230

Broader Connections

• Computational and data limits, not theory, made early portfolio selection unusable. #expands (0.85) — Donald MacKenzie, An Engine, Not a Camera, 230