Screenshots
Technical Analysis Models and Indicators
The following models and indicators are implemented in Optimal Trader. The parameters of the models are specifically optimized for each stock by Optimal Trader when you optimize your portfolio.
Moving Average Convergence/Difference Oscillator, Modified
The MACD indicator is made up of two moving averages, a fast exponential moving average and a slow Hanning filtered moving average. The MACD Oscillator is based on the difference between these two moving averages. The Moving Average Difference Oscillator is a lagging indicator.
Parameter 1, Exponential Moving Average. The value sets the smoothing factor of the moving average. Allowed values are integers larger than zero.
Parameter 2, Hanning Moving Average . The value sets the smoothing factor of the moving average. Allowed values are integers larger than zero.
The values of the parameters are not set in the same scale. If the same value is set for both moving averages the exponential moving average will be slower than the Hanning filtered moving average.
Moving Average Convergence/Difference Oscillator, Adaptive
MACD Adaptive is based on the standard oscillator with the same name, but differs significantly because the fast moving average is adapting to the price's behavior. This is achieved with Optimal Traders in-house developed adaptive moving average, OptAMA. The adaptive characteristics trigger earlier buy and sell signals at turning points of the trend and reduce the sensitivity of the indicator when there is no trend and price movements are small.
Parameter 1, Fast Factor, OptAMA. The value controls how sensitive OptAMA is at major price movements. Smaller values, i.e 0.75, makes OptAMA react faster, than larger values, i.e. 1.5, when price movements are large. Accepted values are integers larger than zero.
Parameter 2, OptAMA Period. The value controls the smoothing (set in days) of the fast adaptive moving average. Accepted values are integers larger than zero.
Parameter 3, Exponential Moving Average. The value controls the smoothing of the slow moving average. Accepted values are integers larger than zero.
The time period of OptAMA and the exponential moving average are defined on the same scale, but cannot be compared to each. OptAMA is often a bit faster than the exponential moving average of the same time period.
Momentum, Classic
Momentum is without doubt the most basic technical indicator. It is based on the difference between the last closing price and the closing price a number of days earlier. In the overanalyzed market situation of today an easy indicator such as Momentum often performs inferior compared to other indicators. Momentum is a lagging indicator.
The price data is smoothed in Optimal Trader to reduce noise. The smoothing and the time period is optimized automatically for every equity.
Parameter 1, Hanning Filtered Moving Average. The value regulates how much the price is smoothed. Allowed values are integers larger than zero.
Parameter 2, Time Delay. The value determines the number of days between the smoothed closing prices when calculating their difference. Allowed values are integers larger than zero.
Momentum, Adaptive
The noise smoothing in the Momentum indicator reduces erroneous signals and makes the signals more robust. But there is a drawback: Trading often come too late. The use of OptAMA reduces this delay considerably.
Parameter 1, Upper Limit of OptAMA. Allowed values are integers larger than zero.
Parameter 2, Lower Limit of OptAMA. Allowed values are numbers larger than or equal to 1.0.
The upper limit regulates the smoothing at relative calm price movements and the lower limit regulates the smoothing when price movements are more dramatic.
Parameter 3, Time Delay. The value determines the number of days between the smoothed closing prices when calculating their difference. Allowed values are integers larger than zero.
Regression Analysis 1
In-house developed indicator with both leading and lagging properties. For those familiar with higher mathematics it can be useful to know that the price is modeled with the help of regression analysis as a polynomial of an order appropriate to the particular price data. The polynomial can give a forecasting of future price values. The model is particular good in finding cycles in price data.
Parameter 1, Hanning Filtered Moving Average. The value regulates how much the price is smoothed. Allowed values are integers larger than zero.
Parameter 2, Regression Data Length. The value determines the time period used for price modeling at a certain time. Allowed values are integers larger than zero.
Parameter 3, Order of Regression Polynomial. Sets the order of the polynomial used to model the price. Allowed values are integers larger than zero.
Parameter 4, Prediction Horizon. Determines how far ahead the model predicts the price to determine which trading signal should be generated. Allowed values are integers larger than or equal to zero.
Regression Analysis 2
Variation of Regression Analysis 1 which sometimes works better.
RSI (Relative Strength Index) 1
Leading indicator developed by Welles Wilder. High RSI values develop when prices have advanced much and low RSI values develop when prices have declined. A buy signal is generated by the RSI curve breaking through the lower limit, and a sell signal is generated when the RSI curve breaks through the upper limit. Because the RSI signal is very noisy this indicator has been much improved with Optimal Traders OptAMA.
Parameter 1, OptAMA Period Length. The value regulates how much the price is smoothed. Allowed values are integers larger than zero.
Parameter 2, RSI Period Length. This value determines the time period used to calculate the RSI signal. Allowed values are integers larger than one.
Parameter 3, RSI Oscillator Lower Limit
Parameter 4, RSI Oscillator Upper Limit
The lower limit determines when the indicator will generate a sell signal and the upper limit determines when the indicator will generate a sell signal.
RSI (Relative Strength Index) 2
Variation of RSI 1 which sometimes produces better results. A buy signal is generated when the RSI curve breaks through the lower limit from below and a sell signal is generated when the RSI curve breaks through the upper limit from above.
Stochastic RSI
Stochastic RSI was developed by Tushar S. Chande and Stanley Kroll and is an indicator of an indicator. It is based on RSI values and the relation between them over a time period to increase the sensitivity of the standard RSI indicator.
Stochastic RSI detects overbought and oversold levels and is thus a leading indicator.
Parameter 1, Period. The value sets the period length for the RSI signal and the Stochastic Oscillator. Allowed values are integers larger than zero.
Parameter 2, OptAMA Period Length. The value regulates the internal smoothing of the indicator. Allowed values are integers larger than zero.
Parameter 3, StochRSI Oscillator Lower Limit
Parameter 4, StochRSI Oscillator Upper Limit
The lower limit determines when the indicator will generate a sell signal and the upper limit determines when the indicator will generate a sell signal.
Parameter 5, Limit for Sideway Trend. If the value is over the threshold, the RSI signal is not trending and no trading signals are emitted.
Stochastic Oscillator 1, Modified
Stochastic Oscillator (developed by George Lane) is an indicator which is constructed to find situations when the stock is considered overbought or oversold and is thus a leading indicator. Buy signales are generated when the Stochastics Curve rises over a threshold and sell signals are generated when the Stochastics Curve falls under that threshold. In Optimal Trader the model is improved to be less sensitive when there is no trend. The indicator is also improved with the help of OptAMA.
Parameter 1, Stochastic Period. This value determines the period used to calculate the Stochastic Signal. Allowed values are integers larger than zero.
Parameter 2, OptAMA Period Length. The value regulates the internal smoothing of the indicator. Allowed values are integers larger than zero.
Parameter 3, Stochastic Oscillator Limit . If the Stochastic Signal is over the threshold buy signals are generated, otherwise sell signals are generated.
Parameter 4, Limit for Sideway Trend. If the value is over the threshold the price is not trending and no trading signals are generated.
Stochastic Oscillator 2, Modified
Variation of Stochastic Oscillator 1 with two limits for the Stochastic Curve. A buy signal is generated when the Stochastic Curve breaks through the lower limit from below and a sell signal is generated when the Stochastic Curve breaks through the upper limit from above.
Parameter 1, Stochastic Period. This value determines the period used to calculate the Stochastic Signal. Allowed values are integers larger than zero.
Parameter 2, OptAMA Period Length.The value regulates the internal smoothing of the indicator. Allowed values are integers larger than zero.
Parameter 3, Stochastic Oscillator Lower Limit
Parameter 4, Stochastic Oscillator Upper Limit
The lower limit determines when the indicator will generate a sell signal and the upper limit determines when the indicator will generate a sell signal.
Parameter 5, Limit for Sideway Trend. If the value is over the threshold the price is not trending and no trading signals are generated.
Stochastics (Stochastic Oscillator) Iinverted
In-house developed variation of the standard Stochastic Oscillator where parts of the model have been changed.
Parabolic SAR (Stop And Reverse)
Parabolic SAR was developed by Welles Wilder and has been named for the patterns the model creates above and under the price curve. The indicator is trend lagging and contains an acceleration factor for the trend to establish. In Optimal Trader Parabolic SAR is presented with crosses above and under the curve. OptAMA is used to eliminate noise efficiently.
Parameter 1, OptAMA Period Length. The value regulates how much the price is smoothed. Allowed values are integers larger than zero.
Parameter 2, Initial Acceleration. Allowed values are numbers larger than or equal to zero.
Parameter 3, Acceleration Step. Allowed values are numbers larger than zero.
Parameter 4, Max Acceleration. Allowed values are numbers larger than zero.
Artificial Neural Network
Parameter 1, Number of Epochs. Determines how much the network will be trained. Allowed values are integers larger than zero.
Parameter 2, Number of Networks. Because every implementation of the neural network produces slightly different results, the results of several networks are combined to produce a more stable result. The value for this parameter sets the number of networks used for the combined result. Allowed values are integers larger than zero.
Parameter 3, Beta Start. A training factor. Allowed values are numbers between 0 and 1.
Parameter 4, Beta End. A training factor. Allowed values are numbers between 0 and 1.
Parameter 5, Momentum. A training factor. Allowed values are numbers between 0 and 1.
Deterministic Training. Neural network results differ somewhat because of random
neural network initialization. Deterministic Training
will generate identical results under the same conditions because
the same initialization is always used.
Disadvantage:
Sometimes the neural network fails in finding a good
solution. In that case you should re-optimize the network.
But if you have selected Deterministic Training, the
results will always be the same.
Combined Analysis Model
The Combined Analysis Model is not a model by itself, but a weighted fusion of the results from the active models.
If more active models are generating buy signals the Strength Signal of the Combined Analysis Model will be stronger.
If the weight of a model is increased, the model will affect the weighted fusion more and have a greater influence on the Strength Signal.
Parameter 1-5, Weight. Sets the weight for each active model. The weight of each model determines the importance of the model. The larger the weight for a specific model is, the more the corresponding model willcontribute to the Strength Signal of the Combined Analysis Model. Allowed values are numbers between 0.0 and 1.0.
Trailing Stop-Loss Indicator
One of the most basic techniques for establishing an appropriate exit point is the trailing-stop technique, which is very popular among new traders. Trailing stop-losses have the best of both worlds: you can protect yourself from losing too much, and also lock in more and more profits at the same time.
What is a trailing stop-loss?
Very simply, the trailing stop is a stop-loss order in which the stop-loss price is set at some fixed percentage below the market price. If the market price rises, the stop-loss price rises proportionately, but if the stock price falls, the stop-loss price doesn't change. When the stock falls and hits the limit, the stop-loss order becomes a market order. A market order instructs your broker to sell immediately at the best possible price. In a volatile market, you may not get the price you wanted, but it should be close.
This technique has several advantages to an investor:
- You set a limit on the maximum possible loss
- You do no set a limit on the maximum possible gain
- You don't need to pay attention to the investment on an ongoing basis. The broker will sell your stock as soon as the limit is hit.
- If your broker allows you to set a trailing buy order you can also set the limit based on the values of this indicator.
Example: A trailing stop-loss order with a 10% limit would kick in at $41.10 if the price is $46 ($46 x 10% = $4.60; $46 - $4.60 = $41.10). If the stock keeps moving up, so will the trailing stop.
- At $47 per share, the trailing stop is $42.30
- At $48 per share, the trailing stop is $43.20
- At $49 per share, the trailing stop is $44.10
- At $50 per share, the trailing stop is $45.00
As long as the stock keeps rising or holds relatively steady, nothing happens. However, if it turns down and hits your trailing stop, your broker will sell the stock. It is important to note, the trailing stop only goes up, it never goes down with a market price. Thus, if the price falls back to $46, the trailing stop will still be $45.00.
Deciding what constitutes an appropriate trailing-stop percentage is perhaps the most difficult aspect of establishing a trailing-stop system for your disciplined trading decisions. The trick is setting the percentage at a level that will pick up a true price drop as opposed to normal daily price fluctuations. Setting your trailing-stop percentage can be done using a relatively vague approach (closer to emotion) or by following a more analytical approach.
Optimal Traders Trailing Stop-Loss Indicator helps you find optimal trailing-stop percentages for your equities based on their historical behaviour.
You can either optimize the limits automatically or test different limits manually and immediately
evaluate the results.
Remember that:
- Because the order is triggered automatically when the stop price is reached, the stop price could be activated by a short-term fluctuation in an equity's price.
- In a fast-moving market, the price at which the trade is executed may be different from the stop price.
Parameter 1, OptAMA Period Length. The value regulates how much the price is smoothed. Allowed values are integers larger than zero. By default, this parameter is disabled and no smoothing is applied when optimizing.
Parameter 2, Sell Limit. This value sets the sell limit in percentage of the highest price since purchase. If the price falls below the limit a sell signal is generated.
Parameter 3, Buy Limit. This value sets the buy limit in percentage of the lowest price since your last selling or since start. If you are not holding the equity and the price rises above the limit a buy signal is generated.
Lock limits to same value. Locks the sell limit and the buy limit to the same value when optimizing.
The Buy Limit can be used if your broker allows you to set a trailing buy order.
If you do not want to use this indicator to find limits for your broker, you may enable the first parameter (adaptive smoothing) and use the full potential of this indicator. Follow the steps below to activate the smoothing parameter:
- Make sure that the Trailing Stop-Loss indicator is one of the five active models. If it is not, click Select Models and select the indicator as one of the active models.
- Click Model Settings and increase the Parameter Accuracy to 25 for the first parameter of the Trailing Stop-Loss indicator.
- In the right tab, select the Trailing Stop-Loss indicator and set the max value of the smoothing parameter to 150.
The steps above ensures that the smoothing factor will be examined with 25 steps from 1 (no smoothing) to 150 (maximum smoothing).
| Copyright © 2005-2008 Optimal Trader |