Archive for April, 2011

Because of the initial random weights, the different runs of the NN backtests give different results. Annoying, but random initial weight is a tool that helps to avoid to stuck in local minima. Let’s suppose we train 21 different NNs with exactly the same parameters. Some of them will be stuck in local minima, some […]

In this post, we look more closely at the trained Encog NN for gaining more insight about the inner workings of the NN. This is what many amateur NN developers usually neglect. However, we think it is cardinal to understand how the NN works. This is why we still use small, simple NNs, to have […]

In this post, we still stick to the most basic case. input: CurrDayChange, output nextDayChange. So far we have backtested our methods with the RUT (Russell 2000 index) only. Our justification is: – RUT is less popular than SPX or DJI in the trader’s community. Therefore, less likely that clever traders/computer bots optimized or removed […]