The goal of this paper is to equip the investor with the tools and understanding necessary to evaluate managed currencies' investments in a meaningful way. It is shown that managed currency funds might exhibit a common factor because most of the trading managers use similar technical forecasts to trigger their positions in the financial markets. Therefore, a dynamic benchmark is built, based on technical trading rules. Using the stochastic properties of trading rules, three simple moving averages are selected and given equal weight. Then the basket of trading rules is applied to a set of currencies. The weighting between currencies is done according to volumes traded on the OTC market as observed through Reuters 2000. Such a dynamic benchmark when adjusted for the leverage and risk-free factors exhibits similar performances, namely returns and volatility, to currency traders' benchmarks. The degree of correlation is high and the tracking error is low. These results might have several implications for institutions wishing to consider managed currency funds. First, the dynamic index might be used as a test of market inefficiencies. Second, the technical index might be used as a benchmark for currency trading advisers. As a whole, it can be seen that managed currencies have been trend-followers because the correlation coefficient between the dynamic index and the currency managers is significantly positive. The dynamic index might well be used to distinguish trend-followers from contrarian and judgemental fund. Finally, the dynamic index might be used as a tool to fulfil market expectations. On the one hand, an investor anticipating trending markets might wish to buy the dynamic index. On the other hand, an investor forecasting range-trading markets might wish to sell the index. In sum, the dynamic index might constitute a new financial product, as well as an appropriate benchmark for managed currencies funds.Benchmark;Currencies;Trend;Trading ;Rules,

No abstract

No abstract

A general framework for analysing trading rules is presented. We discuss different return concepts and different statistical processes for returns. We then concentrate on moving average trading rules and show, in the case of moving average models of length two, closed form expressions for the characteristic function of realized returns when the underlying return process follows a switching Markovian Gaussian process. An example is included which illustrates the technique.Moving Averages, Switching Markov Models, Trading Rules,

No abstract