RupeeCase
Education . Portfolio Theory . 3 of 3

Three rules for honest diversification.

A 5-year backtest in calm markets is the wrong sample for this question. Run these three before you trust the pie chart on your statement.
01

Test correlations in the worst 10 pct of months, not the average month.

The protection you are buying is for those weeks specifically. If a pair sits at rho 0.2 for 90 pct of the sample and rho 0.7 for the worst 10 pct, the second number is the one that pays your rent in a crash. Average-of-history correlation tables are flattering, not honest.

02

Treat any pair with rho above 0.5 as one bet, not two.

Largecap and smallcap. Equity and high yield credit. India and broader emerging markets. Two columns on a factsheet, but one trade in stress. Stress-test rho, not calm rho. Sleeves that fail this test are not diversifiers, they are duplicates with different fee plates.

03

Real diversifiers cost you something in calm years.

If every sleeve in your book is up together every year, you are not diversified. You are levered to the same single regime. Honest diversification means at least one sleeve should be quietly underperforming when the rest are running hot. That sleeve is the premium you are paying for the protection, and it is the one most people sell first.

The diversification benefit you can count on is the one that survives the regime where you actually need it. Everything else was rented, not owned.
Stress-test correlation, drawdown and recovery on the RupeeCase Learn library before you size any sleeve.
rupeecase.com / learn