TradingSolutions Glossary

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

L2 Temporal Criterion

A specific type of criterion that includes the ability to adjust the error based to better model time-related data.

&  For help with neural network components, see the Advanced Topology Settings Dialog help.

Laguarre Memory

A type of memory used in dynamic neural networks. Laguarre memory is similar to gamma memory in that a percentage of each new value is combined with an opposing percentage of the previous value. However, Laguarre memory includes improvements that make the values of later memory taps include more specific information.

&  For help with neural network components, see the Advanced Topology Settings Dialog help.

&  For help with neural network topologies, see the Modify Field Dialog: Prediction Model page help.

Learning Curve

A graph of the difference between the desired output and the output produced by the neural network for each epoch of training.

&  For help analyzing predictions, see Analyzing Predictions.

Learning Rate

A group of values that control the amount by which the training weights in a neural network component will be adjusted based on the backpropagated error. In TradingSolutions, the learning rate is made up of the step size and the momentum.

&  For help with neural network settings, see the Advanced Topology Settings Dialog help.

Limit Order

An order to buy or sell a security at a given price or better. Limit orders have the advantage of naming an advantageous price, but run the risk of not being executed in a timely manner, if at all. Limit orders have lower precedence than market orders.
For example, if the current price of a security is 50 and you would like to enter long, you may place a limit order at 49 in hopes of buying at that price before it goes up.
Limit orders can be simulated during signal analysis during trade entry to help increase profitability.

&  For help using limit orders, see Using Trading Styles.

Local Minima

A reference to one or more local minimums.

Local Minimum

In the performance surface of a neural network, a combination of training weights that produces low errors that are not the lowest possible error, but are surrounded in the graph only by solutions with higher errors. A local minimum causes problems when training a neural network since the training does not know if it is the lowest possible error or which direction to adjust the weights in order to find a lower error.
TradingSolutions uses momentum as part of its learning in order to continue in a positive direction past smaller local minima.

Long Trade

Buying a security at the current price with the intent of selling that security after the price has gone up. In effect, betting that the price will increase.

Losers

Trades that produced no profit or lost money.

Loss

The percentage that the equity has decreased.

Low

The lowest price at which a security traded during a given day.