Asset allocation
Our newsletters use our own proprietary algorithm for making investment recommendations developed by our founder, who is a Ph.D. statistician. This algorithm is a specific implementation of a general idea for making investment decisions called asset allocation. Like other asset allocation methods, our algorithm picks the asset mix that maximizes the expected return while not exceeding a fixed maximum risk. We recalculate the optimal asset mix fairly frequently. This assures that our method responds quickly to changes in the market.
Conceptually, asset allocation is simple. It works as follows.
- Pick a risk level that you are comfortable with. Each one of our newsletters gives recommendations based on two risk levels — conservative and balanced.
- Consider all possible combinations of investments whose risk is below this level.
- Calculate the expected return for each of these investment combinations.
- Pick the combination with the highest expected return.
While the concept is simple, the specifics required to implement the algorithm are not. Here are some issues that must be resolved before asset allocation can be implemented.
- We need specific mathematical definitions of expected return and risk. For example, the mathematical definition of risk that is typically used does not accurately describe how many investors think about risk.
- We need an estimation methodology that takes into account the fact that expected return and risk of each investment can change with time. Not taking this into account leads to a more primitive form of asset allocation called static diversification.
- Finally, we need an algorithm for searching through all the myriads of possible investment mixes to find the one with the highest expected return that does not exceed the predefined maximum risk.
We believe that we have found a good method of implementing asset allocation that addresses all of these issues and more. Here are some features of our method.
- We use our own algorithm, including our own proprietary risk measure. As our algorithm was developed by our founder, who is a Ph.D. statistician, we believe it to be better than many other investment approaches.
- Our algorithm is fully automated. It tells you the exact proportion of your money to put into each investment.
- We do not attempt to predict the future of the economy. We believe that if the economic factors that affect the market are really changing, that will be reflected in the market itself. We thus simply respond to changing market conditions.
- We do not try to pick price lows and highs. We do not time the markets. We simply respond to changing market conditions.
- We do not follow static diversification rules, like "70% stocks / 30% bonds". Our optimal asset mix is updated relatively frequently in response to changes in market conditions. Sometimes, we are diversified among several investments. Sometimes, we allocate all the money into a single investment.