def random_optimise_with_agent_1(backtest, n=20, verbose=False, sort=True):
from Agents.agent_1_simple_macd import SimpleMACDAgent
search_dict = {
'fast_length': [5, 400, int],
'slow_length': [20, 1000, int],
'rets_length': 'slow_length',
'signal_mean_length': [1, 25, int]
}
test_param_balances = random_search_max_expected_return(
SimpleMACDAgent, search_dict, n, backtest, verbose, sort)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def random_optimise_with_agent_2(backtest, n=20, verbose=False, sort=True):
from Agents.agent_2_simple_risk_managed_macd import SimpleRiskMACDAgent
search_dict = {
'fast_length': [5, 400, int],
'slow_length': [20, 1000, int],
'rets_length': 'slow_length',
'signal_mean_length': [1, 25, int],
'stop_loss_scaling': [1.1, 6.0, float],
'take_profit_scaling': [0.1, 6.0, float]
}
test_param_balances = random_search_max_expected_return(
SimpleRiskMACDAgent, search_dict, n, backtest, verbose, sort)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def random_optimise_with_agent_3(backtest, n=20, verbose=False, sort=True):
from Agents.agent_3_ret_bound_risk_macd import RetBoundRiskMACDAgent
search_dict = {
'fast_length': [5, 200, int],
'slow_length': [150, 300, int],
'rets_length': 'slow_length',
'signal_mean_length': [1, 25, int],
'ret_upper_scaling_factor': [0.1, 6.0, float],
'ret_lower_scaling_factor': [0.1, 6.0, float]
}
test_param_balances = random_search_max_expected_return(
RetBoundRiskMACDAgent, search_dict, n, backtest, verbose, sort)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def expicit_optimise_with_agent_1(backtest,
verbose=False, sort=False):
from Agents.agent_1_simple_macd import SimpleMACDAgent
test_cases = {'fast_length': [20, 60],
'slow_length': [40, 120],
'rets_length': 'slow_length'}
test_param_balances = explicit_search_max_expected_return(
SimpleMACDAgent, test_cases,
backtest=backtest,
verbose=verbose,
sort=sort
)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def random_optimise_with_agent_4(backtest, n, verbose=False, sort=True):
from Agents.agent_4_decision_tree import DecisionTreeAgent
search_dict = {
'fast_length': [5, 200, int],
'slow_length': [150, 300, int],
'rets_length': 'slow_length',
'signal_mean_length': [1, 25, int],
'prediction_horizon': [20, 500, int],
'max_depth': [1, 5, int],
'target_profit': [0.01, 0.5, float]
}
test_param_balances = random_search_max_expected_return(
DecisionTreeAgent, search_dict, n, backtest, verbose, sort)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def expicit_optimise_with_agent_3(backtest,
verbose=False, sort=False):
from Agents.agent_3_ret_bound_risk_macd import RetBoundRiskMACDAgent
test_cases = {'fast_length': [120]*2,
'slow_length': [250]*2,
'rets_length': 'slow_length',
'ret_upper_scaling_factor': [2.5, 3.0],
'ret_lower_scaling_factor': [2.5, 3.0]}
test_param_balances = explicit_search_max_expected_return(
RetBoundRiskMACDAgent, test_cases,
backtest=backtest,
verbose=verbose,
sort=sort
)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def expicit_optimise_with_agent_2(backtest,
verbose=False, sort=False):
from Agents.agent_2_simple_risk_managed_macd import SimpleRiskMACDAgent
test_cases = {'fast_length': [120]*2,
'slow_length': [250]*2,
'rets_length': 'slow_length',
'stop_loss_scaling': [1.1, 1.5],
'take_profit_scaling': [3.5, 5.0]}
test_param_balances = explicit_search_max_expected_return(
SimpleRiskMACDAgent, test_cases,
backtest=backtest,
verbose=verbose,
sort=sort
)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
def expicit_optimise_with_agent_4(backtest,
verbose=False, sort=False):
from Agents.agent_4_decision_tree import DecisionTreeAgent
test_cases = {'fast_length': [120]*2,
'slow_length': [250]*2,
'rets_length': 'slow_length',
'prediction_horizon': [200, 250],
'max_depth': [1, 3],
'target_profit': [0.1, 0.2]}
test_param_balances = explicit_search_max_expected_return(
DecisionTreeAgent, test_cases,
backtest=backtest,
verbose=verbose,
sort=sort
)
if verbose:
print_optimisation_outputs(test_param_balances)
return test_param_balances
