Major Features:

1. Models and indicators:
   • Several off the models are classic indicators, but in Optimal Trader they are improved and refined, for instance with adaptive smoothing of the price of the stock. Among the models available you can choose Stochastic Oscillator, StochasticInverse (in-house developed indicator), RSI, Regression Analysis (in several degrees), MACD and Parabolic SAR. Read more about the different models here.
2. Portfolio Scan:
   • With Portfolio Scan you can find winning stocks or funds based on your criteria and neural network forecasting. Optimal Traders Portfolio Scan is a powerful feature because you can combine several criteria, apply a different weight to each criterion and include neural network forecasting.
   • You can also diversify your portfolio with beta values to minimize risk and estimate the predictability of stocks or funds.
   • More info: Portfolio Scan - Pick Winning Stocks.
3. Portfolio Optimization:
   • Portfolio Optimization is a risk management tool which helps you to optimize your portfolio with regard to the risk of the individual assets, the correlation between the assets and the expected returns of your assets. Portfolio Optimization is a powerful and easy to use risk management feature in Optimal Trader which will help you to allocate assets optimally in your portfolio.
   • Check out the illustraded walk-through guide of creating an optimal portfolio.
4. Artificial Neural Network:
   • In addition to the five optimized models there exists another model that always is active: the Top Model.
   • The Top Model is normally constituted of an artificial neural network. Neural networks have a high ability in detecting patterns in stock price development. A neural network can be described as an artificial brain that can be trained to perform certain tasks, for example predicting future stock price values.
5. Combined Analysis Model:
   • You can also choose to let the top model be constituted of a weighted fusion of the active models trading signals.
   • Perhaps you have noticed that some models produce buy signals while at the same time other models may produce sell signals? Which model should you follow? Optimal Trader's Combined Analysis Model merges the buy and sell signals of the active models into a single trading signal. The merging is not static but adapted optimally for each stock. The Combined Analysis Model produces more stable buy and sell signals than single models buy and sell signals.
6. Best Models and Settings:
   • The Best Models and Settings feature returns the models and optimization settings which historically have performed best (from a large database with historical prices) when price behaviour was similar to the current price behaviour.
   • Best Models and Settings also shows you a report with the returns you can expect on average with your current settings and the current price behaviour.
7. Optimization:
   • When performing standard optimization, parameters in the trading system are tested with different values and the combination of values that has yielded the largest profit is thereafter used. Normal optimization produces profitable and stable values using long optimization periods, but there are more careful ways to optimize the parameters.
   • In Optimal Trader you have the possibility of optimizing the models so that they are trying to produce a high and stable profit instead of trying to maximize the total profit. The result is that the probability of profitable future results increases. This option is called Robust Optimization.
   • Another option is to use Robust Parameters which returns parameters that are more stable in the parameter space. Parameter values around optimal solutions are checked and optimal parameter combinations that produce good results when changing the parameters slightly are valued higher than parameter combinations which perform inferior when changing a parameter value slightly. This setting may thus produce more stable results.
   • There are many factors to consider when optimizing parameters and you have to be careful not to overfit the models. In Optimal Trader you can adjust the following optimization factors:
   • How long period backward do you want to use for optimization (optimization period)?
   • Do you want to limit curve fitting of stock data that contain sudden larger price gaps and peaks?
   • Do you want price development closer in time to have a higher influence than price development a long time ago?
   • Does it cost to buy and to sell your share/fund (Brokerage fees, fund fees)?
   • When are your purchases and sells carried out? (default: same day for all equities except for funds)
   • Notice that the goal of optimization is to find optimal parameters, not to give an estimation of future returns. Optimal parameters will almost always produce superior results for the optimized period compared to realistic usage of the system. Estimate more realistic returns with the Best Models and Settings feature or with Back Testing.
   • If you do not want to use optimized parameters can you use Optimal Trader's default parameters which are selected to produce reliable and stable results in most situations, or set the parameters yourself manually.
8. OptAMA - Optimal Trader Adaptive Moving Average:
   • In technical analysis there is an aim to smooth signals in order to reduce noise. Smoothing is achieved by applying a moving average. The advantage is that indicators produce more unambiguous signals, but moving averages also introduce a delay. The more the signals are smoothed, the larger the delay will get. Optimal Trader uses an in-house developed adaptive moving average, OptAMA, that eliminates much of the delay, but nonetheless smoothes the noise efficiently. The result is clearer signals in many models and fewer erroneous signals when the trend is uncertain.
   • Signals come consistently earlier which is important in sudden upward or downward movements, since the largest price differences often happen early in these phases. Read more about OptAMA here.
9. Back Testing:
   • Optimal Trader can make a backward analysis (step-by-step Back Testing) to evaluate the results of following a certain stock with a certain model. The function simulates use of a model through day-by-day optimization a number of days backward in time in order to use the same buy and sell signals you would have gotten if you would have followed the model in reality and optimized it each day.
   • Notice that this is not the same as carrying out a simple back test, i.e. to look on how a certain model would have performed backward in the time with some parameters. It is always easy to find good parameter values for a period afterward, just as it is easy to realize how you should have traded a stock afterward. Step-by-step Back Testing gives you a more reliable indication on a model's strength.
   • One drawback is that Back Testing results are very sensitive to prediction results for rare major price movements. No matter how long the period you are backtesting for, a single 10% investment increase or decrease of a specific day will still affect the overall performance of the complete period with 10%.
   • Another drawback is that the results, which are a measure of the predictiveness, only are based on a single time series.
   • Thus, a more stable analysis is provided with the Best Models and Settings feature where you can select to generate report.