def initialize_portfolio(initialize_variables, init_pull):
pytest.return_dict = calculate_returns(pytest.stock_dict['stock_dict'])
pytest.interest_rates = get_risk_free_rate(pytest.start_date,
pytest.end_date)
pytest.testPortfolio = Portfolio(pytest.start_date, pytest.end_date,
pytest.return_dict, pytest.interest_rates,
pytest.frequency,
pytest.transaction_costs)
return pytest.testPortfolio
def test_dailyrf_rate(self, initialize_variables):
#Configuration
start_date = pytest.start_date
end_date = pytest.end_date
#Nested JSON
interest_rates = get_risk_free_rate(start_date, end_date)
assert abs(
interest_rates['daily_rf_rate'] - .0041
) < .0001 #Known to be approximately .0041 for the month of January 2018
def test_dailyrf_rate(self):
#Configuration
start_date = pd.to_datetime('2018-01-01')
end_date = pd.to_datetime('2018-01-31')
#Nested JSON
interest_rates = get_risk_free_rate(start_date, end_date)
assert abs(
interest_rates['daily_rf_rate'] - .0041
) < .0001 #Known to be approximately .0041 for the month of January 2018
def parse_info():
#check date is valid format
pattern = re.compile(
"^\d{4}\-(0?[1-9]|1[012])\-(0?[1-9]|[12][0-9]|3[01])$")
if not pattern.fullmatch(request.get_json()['start_date']):
abort(400)
if not pattern.fullmatch(request.get_json()['end_date']):
abort(400)
stocks = request.get_json()['assets']
benchmark = request.get_json()['benchmark'] #Needs to be a list
start_date = pd.to_datetime(
request.get_json()['start_date']) #Datetime object
end_date = pd.to_datetime(request.get_json()['end_date'])
frequency = request.get_json()['frequency']
transaction_costs = request.get_json()['transaction_costs'] #0 or 1
#is date valid range?
#Error code 1
if start_date >= end_date:
return jsonify({
"Error_Code": "1",
"Error_Description": "Inverted Date Range"
})
#Used for date interval checking
date_diff = end_date - start_date
#Error code 2
if frequency == "monthly" and date_diff.days > 731: #2 years worth of days
return jsonify({
"Error_Code":
"2",
"Error_Description":
"Too long of a date range (monthly). Please enter a range interval between 4 months and 2 years"
})
#Error code 3
if frequency == "quarterly" and date_diff.days > 1461: #4 years worth of days
return jsonify({
"Error_Code":
"3",
"Error_Description":
"Too long of a date range (quarterly). Please enter a range interval between 1 year and 4 years"
})
#Error code 4
if frequency == "biannual" and date_diff.days > 2922: #4 years worth of days
return jsonify({
"Error_Code":
"4",
"Error_Description":
"Too long of a date range (biannual). Please enter a range interval between 2 year and 8 years"
})
#Error code 5
if frequency == "monthly" and date_diff.days < 120: #4 months worth of days
return jsonify({
"Error_Code":
"5",
"Error_Description":
"Too short of a date range (monthly). Please enter a range interval between 4 months and 2 years"
})
#Error code 6
if frequency == "quarterly" and date_diff.days < 365: #1 year worth of days
return jsonify({
"Error_Code":
"6",
"Error_Description":
"Too short of a date range (quarterly). Please enter a range interval between 1 year and 4 years"
})
#Error code 7
if frequency == "biannual" and date_diff.days < 731: #2 years worth of days
return jsonify({
"Error_Code":
"7",
"Error_Description":
"Too short of a date range (biannual). Please enter a range interval between 2 year and 8 years"
})
#check if is string, make list if string
if isinstance(benchmark, str):
benchmark = [benchmark]
#ensure stocks is list
if not isinstance(stocks, list):
abort(400)
#Initial data pull
results = pull_data(stocks, start_date, end_date)
benchmark_results = pull_data(benchmark, start_date, end_date)
#Get daily and annual risk free rate
interest_rates = get_risk_free_rate(start_date, end_date)
#Calculate log returns
return_dict = calculate_returns(results['stock_dict'])
benchmark_return_dict = calculate_returns(benchmark_results['stock_dict'])
#Output for the portfolio
portfolio = Portfolio(start_date, end_date, return_dict, interest_rates,
frequency, transaction_costs)
output = portfolio.optimize_portfolio()
#Output for the benchmark
benchmark = Benchmark(benchmark_return_dict, benchmark)
benchmark_output = benchmark.form_returns()
#Find matching keys, build joint data structure
common_keys = set(output['cumulative_returns']).intersection(
benchmark_output["benchmark_cumulative_returns"])
intersect_dict = {}
for key in common_keys:
intersect_dict[key] = [
output['cumulative_returns'][key],
benchmark_output["benchmark_cumulative_returns"][key]
]
return jsonify({
"optimized_cumulative_returns":
output['cumulative_returns'],
"optimized_weights":
output['optimized_weights'],
"benchmark_portfolio_intersection":
intersect_dict,
"benchmark_cumulative_returns":
benchmark_output['benchmark_cumulative_returns']
})