New data, derived from a sample of 1000 funds, is now available for the Dynamic Trading Risk Factor. Before going into the performance of the factor, which I will do in a following post, I thought I would discuss some changes to the constituency of the Poor Man's Hedge Fund, our a portfolio designed to replicate the performance of hedge funds by investing in the common stocks of firms which have monthly returns well described by the factor.

The table shows the eight members of the XLF benchmark index that have the highest regression R² onto the dynamic trading risk factor. The membership of the portfolio is unchanged, although there has been some reordering in response to the new data. In particular, JNS has dropped out of the top 5 and been replaced by AMP.

At this time, I also decided to change slightly the relative composition of the portfolio. I have been constructing an equal weighted portfolio with 60% of the capital allocated to the member stocks and 40% to the hedge. From now on, those members will be individually weighted by their particular R² relative to the portfolio mean. The changes are actually slight.

It's now three month's since we started trading a portfolio designed to replicate the performance of hedge funds by picking those stocks with maximum exposure to the dynamic trading risk factor.

The portfolio's performance peaked in early August, and has been in a drawdown since then. It was hit hard on the 1st. September, when all the financial stocks dramatically underperformed the market. However, it is still profitable after fees.

UPDATE 03/15/2010: The performance chart has been replaced with the one for this current year. As you can see, the system is currently in a small drawdown.

I've now been invested personally in the hedged portfolio for A Poor Man's Hedge Fund, and publishing the contituents daily to my page on Twitter. The first two month's daily profit-and-loss is illustrated on the chart below.

Note that this portfolio is hedged against market returns by putting 40% of it's assets into the SDS ultra-short ETF. (This was originally 33%, but increased it to 40% because the out-of-sample regression onto SPY gave a positive coefficient, so I increased the hedge ratio.) The positions taken have a nominal capital of around $450,000 — without leverage these profits are equivalent to an annualized return of 13%.

With an extra month's data for the Dynamic Trading Risk Factor, we can look to see whether there has been any re-ordering of the members of the XLF that are selected for membership of the Poor Man's Hedge Fund.

The above data shows no change to our prior computation, and the membership is still:

1. Invesco plc IVZ
2. Goldman Sachs Group Inc. GS
3. Morgan Stanley MS
4. T Rowe Price Group Inc. TROW
5. Janus Capital Group Inc. JNS
6. Ameriprise Financial Services Inc. AMP
7. Franklin Resources, Inc. BEN
8. Lincoln National Corp. Inc. LNC

Note that Ameriprise is no longer in the top five, having been replaced by Janus.

I introduced the idea of Poor Man's Hedge Fund on 2nd. June. On the 8th., I decided to put the trade on, and have been holding and rebalancing this portfolio since that date. There has been only one modification, which was that I adjusted the universe size when evidence indicated that the portfolio size should probably be a little larger . That change was made on 25th. June.

The performance for the model portfolio traded is illustrated in the chart above. Almost immediately after the initial trades were executed, we were hit by the drop in Ameriprise, when they announced that they would seek $900MM of additional capital, diluting existing shareholders and dominating the performance of the portfolio for the first week. It took the rest of the month to get that money back.

Our original Poor Man's Hedge Fund was built from five stocks. However, evidence presented in a recent post, studying the tracking of portfolios of various sizes, suggests that the right number might be larger. If one looks at the chart in detail, it's probably best to conclude that the optimal size is somewhere in the region 4–10 stocks. Since my prior bias is too include more stocks, thereby acting to reduce the exposure to idiosyncratic factors, I decided to go with the eight stocks that give the best tracking, as measured by the in sample regression R². The data for these eight are listed in the table below.

The added companies are:

1. Janus Capital Group Inc. JNS
2. Franklin Resources Inc. BEN
3. Lincoln National Corp. LNC