Transaction Cost Analysis A-Z

Transaction Cost Analysis A-Z — November 2008

IV. Estimating Transaction Costs with Pre-Trade Analysis

the groundwork for implementation that meets the investor’s goals and respects his preferences. Various sophisticated tools and models are on offer today to this end. Our objective is not to review all of them; it is impossible because there are too many and there is little public information about how they work, essentially for commercial reasons. We intend instead to document the construction of a complete framework for proper forecasting of transaction costs. Accordingly, we have chosen to refer hereafter to the analytical approach proposed by Kissell and Glantz (2003). These authors put in the public domain a considerable amount of technical and very detailed material dealing with the means of modeling and forecasting implicit transaction costs. Their work allows us to illustrate the approach without promoting a particular commercial tool. We will see also that this approach makes it possible to determine the cost profile of any trading strategy through simultaneous estimation of cost and risk, which is the basis for developing optimisation techniques to derive an efficient trading frontier and determine the optimal strategy most closely aligned with the investor’s final objective. (1) Cost structure Kissell and Glantz (2003) build their analytical framework for cost estimation on the following price trajectory formulation: P t = P t − 1 + U t + K t + E t , where P t is the price of trade t , U t is the natural price appreciation from time t- 1 to t , K t is the market impact of trade t and E t is the price volatility with E t ~N(0, σ ² ). In this price trajectory, we can identify price appreciation ( U t ) and timing risk ( E t ) as drivers for opportunity costs as well as

market impact ( K t ). For the latter, we simplify reality since there is no distinction between permanent impact and temporary impact that dissipates over time. From the above cost structure, estimating transaction costs results in making the sum of the forecasted value of each driver: price appreciation, market impact and timing risk. To get such values, we need estimation models that are able to predict cost values when applied to a specific trading strategy. We review below each cost driver and use Kissell and Glantz’s material to show what such models look like and how they deliver estimates. 12 (2) Price appreciation As explained above, price appreciation is the natural price movement of a security. It is sometimes referred to as price trend. It is a truism to say that transaction costs depend on price appreciation: momentum traders tend to incur large transaction costs because they buy when prices have risen and sell when they have fallen. By contrast, contrarian traders buy when prices have fallen and sell when they have risen, so their transaction costs tend to be lower. Taking price appreciation into account thus adds substantial value to the implementation process. To predict the cost of price appreciation, we need first to develop price forecasts over the trading horizon and then determine cost estimates for the specified execution strategy. (a) Price forecast models Both fundamental analysis and technical analysis provide price forecasts in the form of expected short-term or long-term price

12 - We present here a summarised version of the models developed by the authors. For more detail, see Kissell and Glantz (2003).

42

An EDHEC Risk and Asset Management Research Centre Publication