@techreport{Vogel2017, author = {Vogel, Peter}, title = {Investing as Random Trial}, series = {Trading}, journal = {Trading}, number = {1}, publisher = {Hochschule D{\"u}sseldorf}, address = {D{\"u}sseldorf}, issn = {2567-2347}, doi = {10.20385/2567-2347/2017.1}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:due62-opus-11118}, pages = {39}, year = {2017}, abstract = {We introduce an investment algorithm for a market of individual securities. The investment algorithm is derived from constraints depending on investment parameters in order to limit the risk and to take into account an individual investor. One constraint is devoted to trading costs. Purchased securities are selected randomly among securities that meet the buy condition, making trading a random trial. Simulations with historical price data are demonstrated for a simple example: The buy condition is evaluated on the basis of the price relationship for two subsequent trading days and the sales condition is defined by holding securities only for one day. A trading expert evaluates the expected return for the investment algorithm with respect to the random selection. Thus, the expert informs precisely on how many market players perform using the same investment algorithm. Its findings are for a parametrized set of buy conditions simultaneously, which makes a trading expert a valuable tool for theorists as well as for practitioners. In our example, the trading expert demonstrated clearly a significant mean reversion effect for a horizon of one day.}, language = {en} }