Octagon Capital adopts a quantitative approach to investing and is run by Lam Poh Min and Nelson Chia in Singapore. Poh Min has ten years of portfolio management experience. Prior to setting up Octagon, he spent nine years with Singapore's Government Investment Corporation (GIC). He was a senior portfolio manager in the quantitative investment unit. Nelson was a trader at GIC for nine years, working in Singapore, London and New York.
The Octagon Pan Asia Fund is a trend-following, directional long/short equity fund that invests across Asia, including Japan, Australia and India, covering ten markets, 2,000 stocks and across all market capitalisation levels.
The fund was launched in October 2004 and currently has US$1 million under management.
- What are the fund's return and volatility objectives?
We do not set specific return targets but seek to capitalise on the opportunities offered by the market as they present themselves. However, we do control the portfolio risk and generally limit the annualised volatility to 20%. Our backtest suggests that the strategy can generate roughly 20% returns with that kind of volatility.
- Can you briefly explain the most important variables used in your price action trending model? And what are the weightings that you place on each variable?
We analyse price action (i.e. price and volume) from various perspectives. The key variables are strength and quality of trend, long- and short-term trend, relative trend and macro trend. We assign roughly equal weight to the indicators. Given the marked difference between uptrends and downtrends in individual stocks, we also employ slightly different techniques in going long vs short stocks.
- Is this a short-term or long-term model?
It is an intermediate term model, given the non-negligible transactions costs in Asia. However, we generally find ourselves holding on to the winners for a much longer period (2-4 months) vs the losers (2-4 weeks).
- Why did you decide to start with equities and not another asset class?
We believe the techniques we use are relatively portable. After all, trends are trends given that we look at them from a quantitative and statistical standpoint. And we have used successfully many of these techniques in different markets and different asset classes over the years. However we chose to launch an equity long/short fund first. This is because, while trend-following models are quite commonly found in the managed futures or CTA arena, they are more unique in the long/short equity space. Within the Asia long/short space, managers are predominantly fundamentally oriented bottom-up stock pickers. Again we felt that our quantitative approach could offer some useful diversification to investors within the space. Finally, compared to other asset classes, the equity markets represent a much deeper and richer investment universe in Asia and hence the best way to participate in the growth of the continent.
- With the exception of the short side, is your current model the same one that you were using at GIC?
The model is very similar to the trend-following models we were using in GIC. It is designed based on similar investment principles and quantitative methodologies. It is different in that it is applied to a much broader universe of over 2,000 stocks in Asia. This is a meaningful difference because we firmly believe that the larger and the more heterogeneous the investment universe, the more powerful the model is in picking the best independent opportunities.
- How have you changed the long-only model used at GIC to incorporate shorting; will you incorporate borrowing costs and availability in the new model?
Although we do not short individual stocks in GIC, we do express negative views by way of underweighting relative to a benchmark or using derivatives. Hence the model has not been changed in any significant way to incorporate shorting of individual stocks. On the issue of costs and availability of borrow, given its dynamic nature and poor data availability, we do not feel there is a satisfactory way to quantify them at the moment. As such, we take them into account through our qualitative overlay.
- How much will the qualitative overlay play in stock selection?
We do not override our model. For instance, we will never short a stock if the model says it is bullish. However, we do employ a qualitative overlay as it relates to implementation, particularly on the short side. This is relevant especially in Asia given the technical constraints. You can say we use our qualitative judgment to find the best way to implement and to express the view of the model.
- When will you use equity derivatives and how much of the portfolio will be allocated to those instruments?
We do not use futures to take directional bets in the markets, but may use them sparingly for hedging purposes. However, we do rely on synthetic equity swaps to get access to certain markets where it may be more efficient to do so from an implementation viewpoint.
- What sectors will you focus on and are there any sectors which you will not cover?
We adopt a very broad definition of Asia. Our investment universe currently consists of ten markets, namely Japan, Australia, India, Hong Kong, Taiwan, Korea, Singapore, Thailand, Malaysia and Indonesia. Given the portability of our strategy, we are constantly looking to further richen our investment universe. The next market that will likely be added will be China, which we believe will be significant going forward. We do not "focus" on any particular sector or country but will gravitate to where the best opportunities are.
- In countries where it is difficult or expensive to short, will you use index futures as hedges?
Yes, we may do that.
- What will be your stop-loss policy for when trends turn? How will you manage and monitor the downside risks involved in this strategy?
We have strict stop-losses for all our stocks, long and short. However, we do not employ fixed mechanical stop-losses of say 10-20% from point of entry. Rather, the stop level for each stock is identified taking into account its inherent volatility and logical points of support and resistance. Ultimately, our bets are sized such that each position does not cause more than 0.5% of damage to the portfolio NAV should the stop be triggered.
- How much assets do you currently have and what are your market travelling plans?
We launched the fund with seed capital of roughly US$1 million from the partners and a few high net worth investors. So far we have focused on operations and track record but plan to embark on a formal marketing trip in the first half of next year.