Tuesday, January 31, 2006

Portfolio January 2006

I couldn’t have planned a better debut month for this blog. The PM sector was on fire last month. The energies gave a robust showing as well. From the day portfolio tracking started (January 11), the actively managed accounts returned 9.2% and the overall portfolio 5.8%. The rally that opened the year actually added another 4% to my overall portfolio, making this January one of the best months I can remember. Without further ado, this is how the portfolio looks like at the end of January.

EGO was the only buy last month (at $4.82, see Stochastics at Work). There was no sell. The table below summarizes the monthly returns.

How the returns are calculated

The monthly returns are not simply monthly changes in the portfolio value, instead they are calculated net of contributions. Otherwise, the figures will be distorted by our cash flow situation and won’t be a true reflection of my investment prowess. After all, this blog is about investing, not personal finance. Specifically,

Gain/loss $ (month n) = Portfolio value (month n+1) – Portfolio value (month n) – Contributions (month n)

Gain/loss % (month n) = Gain/loss $ (month n) / [Portfolio value (month n) + Contributions (month n)/2]

The returns calculated this way are called time weighted returns (TWR). The second formula assumes the contributions are made in the middle of the month which is known in the jargon as the Midpoint Dietz Method. Note the contributions can be a negative number, reflecting a net withdrawal. The cumulative returns are calculated as follows:

Cumulative return (month 1) = Gain/loss (month 1)

Cumulative return (month n) = [1 + Cumulative return (month n-1)] [1 + Gain/loss (month n)] – 1, for n > 1

This is known as chain-linking TWRs, or taking the geometric mean. The implicit assumption of this calculation is that all gains are reinvested.

All returns are pre-tax. Taxes will be paid either by separate funds, or if withdrawn from the portfolio, accounted for in the monthly contributions.

These formulas and much more can be found here.