def format_reconcile_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are different segments
:return:
"""
formatted_output=[]
formatted_output.append(header("Reconcile report produced on %s" % (str(datetime.datetime.now()))))
table1_df = results_object['positions_mine']
table1 = table('Positions in DB', table1_df)
formatted_output.append(table1)
table2_df = results_object['positions_ib']
table2 = table('Positions broker', table2_df)
formatted_output.append(table2)
body_text("Position breaks %s" % results_object['position_breaks'])
table3_df = results_object['trades_mine']
table3 = table('Trades in DB', table3_df)
formatted_output.append(table3)
table4_df = results_object['trades_ib']
table4 = table('Trades from broker', table4_df)
formatted_output.append(table4)
formatted_output.append(header("END OF STATUS REPORT"))
return formatted_output
def format_status_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are different segments
:return:
"""
formatted_output = []
formatted_output.append(
header("Status report produced on %s" %
(str(datetime.datetime.now()))))
table1_df = results_object['process']
table1 = table('Status of processses', table1_df)
formatted_output.append(table1)
table2_df = results_object['method']
table2 = table('Status of methods', table2_df)
formatted_output.append(table2)
table3_df = results_object['price']
table3 = table('Status of adjusted price / FX price collection', table3_df)
formatted_output.append(table3)
table4_df = results_object['position']
table4 = table('Status of optimal position generation', table4_df)
formatted_output.append(table4)
formatted_output.append(header("END OF STATUS REPORT"))
return formatted_output
def format_pandl_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are instruments, contains roll information
:return:
"""
formatted_output = []
formatted_output.append(
header(
"P&L report produced on %s from %s to %s"
% (
str(datetime.datetime.now()),
str(results_object.start_date),
str(results_object.end_date),
)
)
)
formatted_output.append(
body_text("Total p&l is %.3f%%" % results_object.total_capital_pandl)
)
table1_df = results_object.pandl_for_instruments_across_strategies
table1_df = table1_df.round(2)
table1 = table("P&L by instrument for all strategies", table1_df)
formatted_output.append(table1)
formatted_output.append(
body_text("Total futures p&l is %.3f%%" % results_object.futures_total)
)
formatted_output.append(
body_text("Residual p&l is %.3f%%" % results_object.residual)
)
table2_df = results_object.strategies
table2_df = table2_df.round(2)
table2 = table("P&L by strategy", table2_df)
formatted_output.append(table2)
table3_df = results_object.sector_pandl
table3_df = table3_df.round(2)
table3 = table("P&L by asset class", table3_df)
formatted_output.append(table3)
formatted_output.append(header("END OF P&L REPORT"))
return formatted_output
def format_status_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are different segments
:return:
"""
formatted_output = []
formatted_output.append(
header("Status report produced on %s" %
(str(datetime.datetime.now()))))
table1_df = results_object["process"]
table1 = table("Config for process control", table1_df)
formatted_output.append(table1)
table1a_df = results_object["process2"]
table1a = table("Status of process control", table1a_df)
formatted_output.append(table1a)
table1b_df = results_object["process3"]
table1b = table("Status of process control", table1b_df)
formatted_output.append(table1b)
table2_df = results_object["method"]
table2 = table("Status of methods", table2_df)
formatted_output.append(table2)
table3_df = results_object["price"]
table3 = table("Status of adjusted price / FX price collection", table3_df)
formatted_output.append(table3)
table4_df = results_object["position"]
table4 = table("Status of optimal position generation", table4_df)
formatted_output.append(table4)
table5_df = results_object["limits"]
table5 = table("Status of trade limits", table5_df)
formatted_output.append(table5)
table6_df = results_object["position_limits"]
table6 = table("Status of position limits", table6_df)
formatted_output.append(table6)
table7_df = results_object["overrides"]
table7 = table("Status of overrides", table7_df)
formatted_output.append(table7)
text1 = body_text(results_object["locks"])
formatted_output.append(text1)
formatted_output.append(header("END OF STATUS REPORT"))
return formatted_output
def format_reconcile_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are different segments
:return:
"""
formatted_output = []
formatted_output.append(
header(
"Reconcile report produced on %s" %
(str(
datetime.datetime.now()))))
table0_df = results_object["positions_optimal"]
table0 = table("Optimal versus actual positions", table0_df)
formatted_output.append(table0)
table1_df = results_object["positions_mine"]
table1 = table("Positions in DB", table1_df)
formatted_output.append(table1)
table2_df = results_object["positions_ib"]
table2 = table("Positions broker", table2_df)
formatted_output.append(table2)
text1 = body_text(results_object["position_breaks"])
formatted_output.append(text1)
table3_df = results_object["trades_mine"]
table3 = table("Trades in DB", table3_df)
formatted_output.append(table3)
table4_df = results_object["trades_ib"]
table4 = table("Trades from broker", table4_df)
formatted_output.append(table4)
formatted_output.append(header("END OF STATUS REPORT"))
return formatted_output
def format_costs_data(costs_report_data: dict) -> list:
formatted_output = []
formatted_output.append(
header("Costs report produced on %s from %s to %s" %
(str(datetime.datetime.now()), costs_report_data['start_date'],
costs_report_data['end_date'])))
formatted_output.append(body_text("* indicates currently held position"))
table1_df = costs_report_data['combined_df_costs']
table1 = table("Check of slippage", table1_df)
formatted_output.append(table1)
table2_df = costs_report_data['table_of_SR_costs']
table2 = table(
"SR costs (using stored slippage): more than 0.01 means panic",
table2_df)
formatted_output.append(table2)
return formatted_output
def format_liquidity_data(liquidity_report_data: dict) -> list:
formatted_output = []
all_liquidity_df = liquidity_report_data['all_liquidity_df']
formatted_output.append(
header("Liquidity report produced on %s" %
(str(datetime.datetime.now()))))
formatted_output.append(body_text("* indicates currently held position"))
table1_df = all_liquidity_df.sort_values("contracts")
table1 = table(
" Sorted by contracts: Less than 100 contracts a day is a problem",
table1_df)
formatted_output.append(table1)
table2_df = all_liquidity_df.sort_values("risk")
table2 = table(
"Sorted by risk: Less than $1.5 million of risk per day is a problem",
table2_df)
formatted_output.append(table2)
return formatted_output
def format_risk_report(results_dict):
"""
Put the results into a printable format
:param results_dict: dict of risk tables
:return:
"""
formatted_output = []
formatted_output.append(
header("Risk report produced on %s" % str(datetime.datetime.now())))
result1 = results_dict['portfolio_risk_total'] * 100
result1_text = body_text(
"Total risk across all strategies, annualised percentage %.1f" %
result1)
formatted_output.append(result1_text)
table2_df = results_dict['strategy_risk'] * 100
table2_df = table2_df.round(1)
table2 = table("Risk per strategy, annualised percentage", table2_df)
formatted_output.append(table2)
table3_df = results_dict['instrument_risk_data']
table3_df = table3_df.round(1)
table3 = table("Instrument risk", table3_df)
formatted_output.append(table3)
table4_df = results_dict['corr_data']
table4_df = table4_df.round(2)
table4 = table("Correlations", table4_df)
formatted_output.append(table4)
formatted_output.append(header("END OF RISK REPORT"))
return formatted_output
def format_trades_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are different segments
:return:
"""
formatted_output = []
formatted_output.append(
header("Trades report produced on %s" %
(str(datetime.datetime.now()))))
table1_df = results_object['broker_orders']
table1 = table('Broker orders', table1_df)
formatted_output.append(table1)
return formatted_output
def format_trades_data(results_object):
"""
Put the results into a printable format
:param results_dict: dict, keys are different segments
:return:
"""
formatted_output = []
formatted_output.append(
header("Trades report produced on %s" %
(str(datetime.datetime.now()))))
if len(results_object["overview"]) == 0:
formatted_output.append(body_text("No trades in relevant period"))
return formatted_output
table1_df = results_object["overview"]
table1 = table("Broker orders", table1_df)
formatted_output.append(table1)
table2_df = results_object["delays"]
table2 = table("Delays", table2_df)
formatted_output.append(table2)
table3_df = results_object["raw_slippage"]
table3 = table("Slippage (ticks per lot)", table3_df)
formatted_output.append(table3)
table4_df = results_object["vol_slippage"]
table4 = table("Slippage (normalised by annual vol, BP of annual SR)",
table4_df)
formatted_output.append(table4)
table5_df = results_object["cash_slippage"]
table5 = table("Slippage (In base currency)", table5_df)
formatted_output.append(table5)
summary_results_dict = results_object["summary_dict"]
for summary_table_name, summary_table_item in summary_results_dict.items():
summary_table = table("Summary %s" % summary_table_name,
summary_table_item)
formatted_output.append(summary_table)
return formatted_output
def report_system_dynamic(data: dataBlob, backtest: interactiveBacktest):
format_output = []
strategy_name = backtest.strategy_name
timestamp = backtest.timestamp
optimal_positions_df = get_optimal_positions_table_as_df(
data=data, strategy_name=backtest.strategy_name)
optimal_positions_table = table("Optimal positions", optimal_positions_df)
format_output.append(optimal_positions_table)
report_header = header(
"Strategy report for %s backtest timestamp %s produced at %s" %
(strategy_name, timestamp, str(datetime.datetime.now())))
format_output.append(report_header)
format_output = report_system_classic_no_header_or_footer(
data, backtest=backtest, format_output=format_output)
format_output.append(body_text("End of report for %s" % strategy_name))
return format_output
def report_system_classic(data, data_backtest):
"""
:param strategy_name: str
:param data: dataBlob
:param data_backtest: dataBacktest object populated with a specific backtest
:return: list of report format type objects
"""
strategy_name = data_backtest.strategy_name
format_output = []
report_header = header(
"Strategy report for %s backtest timestamp %s produced at %s" %
(strategy_name, data_backtest.timestamp, str(
datetime.datetime.now())))
format_output.append(report_header)
unweighted_forecasts_df = get_forecast_matrix(
data_backtest,
stage_name="forecastScaleCap",
method_name="get_capped_forecast")
unweighted_forecasts_df_rounded = unweighted_forecasts_df.round(1)
unweighted_forecasts_table = table(
"Unweighted forecasts", unweighted_forecasts_df_rounded
)
format_output.append(unweighted_forecasts_table)
# Forecast weights
forecast_weights_df = get_forecast_matrix_over_code(
data_backtest, stage_name="combForecast", method_name="get_forecast_weights")
forecast_weights_df_as_perc = forecast_weights_df * 100
forecast_weights_df_as_perc_rounded = forecast_weights_df_as_perc.round(1)
forecast_weights_table = table(
"Forecast weights", forecast_weights_df_as_perc_rounded
)
format_output.append(forecast_weights_table)
# Weighted forecast
weighted_forecasts_df = forecast_weights_df * unweighted_forecasts_df
weighted_forecast_rounded = weighted_forecasts_df.round(1)
weighted_forecast_table = table(
"Weighted forecasts",
weighted_forecast_rounded)
format_output.append(weighted_forecast_table)
# Cash target
cash_target_dict = data_backtest.system.positionSize.get_daily_cash_vol_target()
cash_target_text = body_text(
"\nVol target calculation %s\n" %
cash_target_dict)
format_output.append(cash_target_text)
# Vol calc
vol_calc_df = get_stage_breakdown_over_codes(
data_backtest,
method_list=[
daily_returns_vol,
daily_denom_price,
rawdata_daily_perc_vol],
)
vol_calc_df["annual % vol"] = vol_calc_df["Daily % vol"] * \
ROOT_BDAYS_INYEAR
vol_calc_df_rounded = vol_calc_df.round(4)
vol_calc_table = table("Vol calculation", vol_calc_df_rounded)
format_output.append(vol_calc_table)
# Subsystem position table
subystem_positions_df = get_stage_breakdown_over_codes(
data_backtest,
method_list=[
get_block_value,
get_price_volatility,
get_instrument_ccy_vol,
get_fx_rate,
get_instrument_value_vol,
get_daily_cash_vol_target,
get_vol_scalar,
get_combined_forecast,
get_subsystem_position,
],
)
subystem_positions_df_rounded = subystem_positions_df.round(2)
subystem_positions_table = table(
"Subsystem position", subystem_positions_df_rounded
)
format_output.append(subystem_positions_table)
# Portfolio position table: ss position, instr weight, IDM, position
# required
portfolio_positions_df = get_stage_breakdown_over_codes(
data_backtest,
method_list=[
get_subsystem_position,
get_instrument_weights,
get_idm,
get_required_portfolio_position,
],
)
portfolio_positions_df_rounded = portfolio_positions_df.round(3)
portfolio_positions_table = table(
"Portfolio positions", portfolio_positions_df_rounded
)
format_output.append(portfolio_positions_table)
# position diags
position_diags_df = calc_position_diags(portfolio_positions_df, subystem_positions_df)
position_diags_df_rounded = position_diags_df.round(2)
position_diags_table = table("Position diags", position_diags_df_rounded)
format_output.append(position_diags_table)
# Position vs buffer table: position required, buffers, actual position
versus_buffers_df = get_stage_breakdown_over_codes(
data_backtest,
method_list=[
get_required_portfolio_position,
get_lower_buffer,
get_upper_buffer,
],
)
instrument_code_list = versus_buffers_df.index
timestamp_positions = get_position_at_timestamp_df_for_instrument_code_list(
data_backtest, data, instrument_code_list)
current_positions = get_current_position_df_for_instrument_code_list(
data_backtest, data, instrument_code_list
)
versus_buffers_and_positions_df = pd.concat(
[versus_buffers_df, timestamp_positions, current_positions], axis=1
)
versus_buffers_and_positions_df_rounded = versus_buffers_and_positions_df.round(
1)
versus_buffers_and_positions_table = table(
"Positions vs buffers", versus_buffers_and_positions_df_rounded
)
format_output.append(versus_buffers_and_positions_table)
format_output.append(body_text("End of report for %s" % strategy_name))
return format_output
def format_roll_data_for_instrument(results_dict):
"""
Put the results into a printable format
:param results_dict: dict, keys are instruments, contains roll information
:return:
"""
instrument_codes = list(results_dict.keys())
formatted_output = []
formatted_output.append(
header("Roll status report produced on %s" %
str(datetime.datetime.now())))
table1_df = pd.DataFrame(
dict(
Status=[results_dict[code]["status"] for code in instrument_codes],
Roll_exp=[
results_dict[code]["roll_expiry"] for code in instrument_codes
],
Prc_exp=[
results_dict[code]["price_expiry"] for code in instrument_codes
],
Crry_exp=[
results_dict[code]["carry_expiry"] for code in instrument_codes
],
),
index=instrument_codes,
)
# sort by time to theoretical roll, and apply same sort order for all
# tables
table1_df = table1_df.sort_values("Roll_exp")
instrument_codes = list(table1_df.index)
table1 = table("Status and time to roll in days", table1_df)
formatted_output.append(table1)
formatted_output.append(
body_text(
"Roll_exp is days until preferred roll set by roll parameters. Prc_exp is days until price contract expires, "
"Crry_exp is days until carry contract expires"))
# will always be 6 wide
width_contract_columns = len(
results_dict[instrument_codes[0]]["contract_labels"])
table2_dict = {}
for col_number in range(width_contract_columns):
table2_dict["C%d" % col_number] = [
str(results_dict[code]["contract_labels"][col_number])
for code in instrument_codes
]
table2_df = pd.DataFrame(table2_dict, index=instrument_codes)
table2 = table("List of contracts", table2_df)
formatted_output.append(table2)
formatted_output.append(body_text("Suffix: p=price, f=forward, c=carry"))
table2b_dict = {}
for col_number in range(width_contract_columns):
table2b_dict["Pos%d" % col_number] = [
results_dict[code]["positions"][col_number]
for code in instrument_codes
]
table2b_df = pd.DataFrame(table2b_dict, index=instrument_codes)
table2b = table("Positions", table2b_df)
formatted_output.append(table2b)
table3_dict = {}
for col_number in range(width_contract_columns):
table3_dict["V%d" % col_number] = [
results_dict[code]["volumes"][col_number]
for code in instrument_codes
]
table3_df = pd.DataFrame(table3_dict, index=instrument_codes)
table3_df = table3_df.round(2)
table3 = table("Relative volumes", table3_df)
formatted_output.append(table3)
formatted_output.append(
body_text(
"Contract volumes over recent days, normalised so largest volume is 1.0"
))
formatted_output.append(header("END OF ROLL REPORT"))
return formatted_output
