Genetic Optimizer: Code Converter
EasyLanguage Code Converter
Run Start menu -> Program -> TradeSmart -> TS GO -> Code Converter
Step 1: Initial Signal
Select the Template that you will be using for
converting signal and copy code of your signal from
EasyLanguage.
For Template determination see Template Description.
Also you can import the signal from text file, click “Open from file” button
and select file.
Parameters of initial signal will be transformed into genes in next step.
Please bear in mind that if you want values of variables that you use in your
signal to be converted into genes for genetic optimization you need to place
them in your signal under Inputs before conversion. You can also edit your
signal later.
Attention! Before copying signal please check if it verifies in EasyLanguage.
Step 2: Genes & Chromosomes
Setup the range of optimization of parameters, minimal step of optimization and
belonging to chromosomes. During this stage parameters from your signal are
being taken and transformed into
genes. All genes
are tied to one
chromosome on default.
Gene – one of parameters in system that can be optimized.
Name – name of gene.
Min – minimum value of gene.
Max – maximum value of gene.
Increment – value increase (step), if = 0 then any values in set range
can be used.
Chromosome – group of genes (parameters) that have logical link for
example parameters of one type of entry or one signal. This group takes part
in hybridization, it is population of parameters that can migrate amongst
samples. That way successful combinations of genes get chances of surveillance
and multiplication. It’s purposive to combine linked parameters, for example
parameters of one signal.
Attention! Only Numeric and Float parameters can be optimized.
String (text) and
TrueFalse parameters cannot be optimized and they will be placed in
Inputs.
You can add your comment in Comment field. It will be placed in signal next in
parenthesis.
Step 3: Inputs Settings
Indicate initial values of parameters that will be specified in Inputs. First parameters are supposed to be for
TSGO working mode management. The rest of
parameters that are not placed in genes list also will be placed here.
Gen – counter of iterations (runs) of your system. After conversion of
the signal, setup optimization in TradeStation for only this parameter from 1
to number of runs (for example 100) and run the optimization. TSGO will
perform optimization of other parameters automatically and after completing the
process it will present optimization report.
Attention! Gen always should be set from 1 to desired number of
iterations of optimization with step 1. Negative and fraction numbers are not
allowed.
ShowInd - (Individ) number of sample in population from which value of
gene is read.
1 - the best sample.
2 - next sample.
ModeTSGO - optimization mode.
0 - deletes old population and begins from empty population.
1 - recalculates old population and continues optimization.
2 - continues optimization without recalculation of old population.
Population - Sets up the size of population.
Size of population in current version is limited from 10 to 1000 (Free version
from 10 to 100).
FreshBlood - Sets up value of factor "fresh blood" in population.
This function increases effective size of population prohibiting possibilities
of leaders domination and giving chance to other candidates. Thus optimization
is accelerated using varying genes bank and avoiding over fitting the system.
Parameters:
FB_factor - value of factor.
=0 - function is turned off, work is same as it was in previous version.
>0 - additional mixing of population is turned on, typical value is 1.
MyReportName – Name of report. As default report will be saved:
…Program Files\TradeSmart\TS GO\Reports
Step 4: Performance Criteria
Select performance criteria that will be showed
in report after optimization.
You can select several, all, one or none criteria.
Selection of signal valuation (Fitness) will be
offered in next stage.
Step 5: Fitness Criteria
Fitness selection (suitability criteria).
You can select one of available criteria or indicate your own in line “Custom”.
Fitness – criteria of applicability, suitability to sample, targeted
function. It is number connected to trading system using which we inform
genetic optimizer how well this trading system satisfy us.
The bigger it is the better is sample. In the simplest case it can be NetProfit of
ProfitFactor, like for example in optimization
using TradeStation. In TSGO user can determine by himself what is
“SURVIVAL” of sample in population of trading system.
In fact it is possible using determination of criteria of optimization to get
system that will show steady results.
Step 6: Finished Code
Copy the gained code after conversion of signal into
EasyLanguage or save it as separate file.
After copying it to EasyLanguage, verify the
signal and run your strategy with this signal.
To do genetic optimization you need to set parameter Gen of your system from 1
to number of runs you want, for example 100. After doing that you can run the
optimization. When it is finished strategy with the best Fitness will be
chosen. You can also see the report of optimization in TS Genetic Optimization
Viewer.
Settings
Please select the necessary Settings
Auto Back-Up saved in Installation directory, for example:
… Program Files\TradeSmart\TS GO\backup.txt
If You need to open Back-Up file select “Show BackUp” in left menu.
If you have encountered any problems converting signal please send us initial
signal, converted signal and description of problem to
feedback form >>>
Or you can write about it at our support forum:
http://support.tsresearchgroup.com
|
