- 10 years old RCC slab load bearing
- Unit Length Vector and Dot Product
- Is $[0,1]$ a countable disjoint union of closed sets?
- How I show that $(b-a)^p \equiv b^p-a^p \mod{p}$?
- Solution of Weighted Nuclear Norm and Total Variation regularizer combined
- Isn't basic open cover same as an open cover?
- Covariance of a Markov chain
- Inverse z transform - help understand left sided region
- Is $L=\{w_1 c w_2\mid w_1,\ w_2\in\{a,b\}^*,\ w_1≠w_2\}$ context free?
- $\mathbb{E}[X_1|X_2]$ of permutation on $9$ elements
- Is $\iota_*:H_1(M-K) \to H_1(M)$ surjective if $K$ is an embedded loop which is null-homotopic?
- Compact embedding of $W^{k,p}(\Omega)$ into $W^{k-1,p*}(\Omega)$, where $p^*$ is the Sobolev conjugate
- Are higher-order Bézier curves envelopes?
- If $A$ is a $2\times 2$ matrix over $\Bbb R$ with $\det(A+I)=1+\det A$ then which is correct:
- An integral involving Bessel function of the first kind of the Sonine-Gegenbauer sort
- An urn with three kinds of balls… and a weird constraint!
- For what n the following inequality holds?
- Jacobian ideal of $f$ and normality
- Maximizing a function of eigenvalue
- Optimizing a Set of Matrix Equations

# Normalize Returns in PCA Hedging?

Lets say I want to form a portfolio of $N$ correlated instruments that minimizes the variance of my daily PnL, given by

$$

PnL = \sum_{i=1}^N h_i \Delta S_i

$$

where $h_i$ are the number of units of instrument $i$ and $\Delta S_i$ is the daily difference in price.

To minimize the variance, I find the covariance matrix of $\{\Delta S_i\}_i$ and perform PCA on this covariance matrix. I then take the eigenvector with smallest eigenvalue and use the entries as my hedging ratios.

My question is, if the daily difference in instruments are vastly different, say on average $\Delta S_1 = \$10$ while $\Delta S_2 = \$0.1$, does it make sense to normalize these before performing PCA?

I an inclined not to because they are in the same units, and I think it makes sense for instruments with a larger variance to contribute more; however, I have always thought I needed to normalize the data before performing PCA, so I am hesitant to do otherwise now.