Archive for May, 2011

We decided to run various combinations of the ensembling when the participating members should all agree on the next day Up or Down decision. Obviously, if not all of them agrees, the strategy stays in cash. Parameters: NensembleGroupMembers = 5; int nTest = 5; Note that we also run the NensembleGroupMembers = 1; experiments (that […]


This is a 24 year backtest. Our framework is best suited to backtest the variation of 2 parameters. That suits well to the human visual system that can spot ‘interesting’ areas mostly in 2D (as a 2D function projection to the third dimension). We have chosen the number of neurons (nNeurons) and the max epoch […]


Is success of the currDayChange input due to the lucky selection of 200 days lookback and the general bullishness/bearishness in that period? Let’s suppose a simple case: In the last 200 days, the average of nextDay%Change of the Up and also the Down days were positive. For example, because we are in a generally bullish […]


The Neural Network FAQ is an immense source of information. In a chapter about normalization, it can be read that removing outliers is not the only solution to attack the outlier contamination. Clipping outliers to specific thresholds is also viable. An obvious idea is to use the SD of the training set to determine the […]


We introduce a new performance measure that scores how consistent, how smooth the profit of the strategy is. In some sense it measures the risk. We measure it on the $PV (Portfolio Value) chart. On the first day, the Portfolio Value is $1. Methods for measuring risk: 1. Usually, a highly variable PV has a […]


Can the Encog NN learn deterministic or non-deterministic, but predictable patterns? It should, but to be sure, we investigate this in this post. When we test these artificially created patters, there is no point using the gCAGR as a performance measure (as it can go to infinity in the good case), so we use the […]