folder_name = r"results\experiment-4"
# ==============================================================================
# === train gradient models
rate = 0.02
drift = 0.05
number_of_tests = 2**7
test_drivers = []
for num in range(number_of_tests):
print(f"dimension {dimension} at test {num + 1} ".ljust(80, "="), end="")
start = perf_counter()
volatility = tf.random.uniform((dimension, ), 0.2, 0.3)
init_instruments, init_numeraire, book = random_books.random_empty_book(
13 / 52, dimension, rate, drift, volatility, num)
# random_books.add_butterfly(init_instruments, book, spread=20)
random_books.add_calls(init_instruments, book)
warmup_driver = gradient_driver.GradientDriver(
timesteps=timesteps,
frequency=0,
init_instruments=init_instruments,
init_numeraire=init_numeraire,
book=book,
learning_rate_min=1e-7,
learning_rate_max=1e-2)
warmup_driver.set_exploration(100.0, 15.0)
warmup_driver.add_testcase(name="value network",
import sys
import os
from time import perf_counter
import utils
import derivatives
import random_books
import gradient_models
import gradient_driver
import hedge_models
tf.get_logger().setLevel('ERROR')
timesteps = 13
init_instruments, init_numeraire, book = random_books.random_empty_book(
timesteps / 52, 1, 0.02, 0.05, 0.2)
init_numeraire = tf.ones_like(init_numeraire)
spread = 10
itm = derivatives.PutCall(book.maturity, init_instruments - spread / 2,
book.instrument_simulator.rate,
book.instrument_simulator.volatility, 1)
atm = derivatives.PutCall(book.maturity, init_instruments,
book.instrument_simulator.rate,
book.instrument_simulator.volatility, 1)
otm = derivatives.PutCall(book.maturity, init_instruments + spread / 2,
book.instrument_simulator.rate,
book.instrument_simulator.volatility, 1)
book.add_derivative(itm, 0, 1)
book.add_derivative(atm, 0, -2)
volatility = 0.2
train_size, test_size, timesteps = int(2**18), int(2**18), 14
alpha = 0.95
dimension = 1
frequency = 4
units = 15
layers = 4
activation = tf.keras.activations.softplus
num_trials = 8
lst_of_drivers = []
risk_measure = partial(hedge_models.ExpectedShortfall, alpha=alpha)
init_instruments, init_numeraire, book = random_books.random_empty_book(
timesteps / 250, dimension, rate, drift, volatility, 69)
random_books.add_rko(init_instruments, book, 5)
driver = utils.HedgeDriver(timesteps=timesteps,
frequency=frequency,
init_instruments=init_instruments,
init_numeraire=init_numeraire,
book=book,
cost=1 / 100 if cost else None,
risk_neutral=not cost,
learning_rate=1e-1)
driver.add_testcase("deep network",
hedge_models.NeuralHedge(
timesteps=timesteps,
instrument_dim=book.instrument_dim,
import hedge_models
import sys
# ==============================================================================
# === hyperparameters
train_size, test_size, timesteps = int(2**2), int(2**14), 7
hedge_multiplier = 5 * int(sys.argv[1])
frequency = 0 # only > 0 for continuous, else 0
alpha = 0.95
folder_name = r"figures\delta-pnl-plots"
case_name = "call"
# ==============================================================================
# === sample data
init_instruments, init_numeraire, book = random_books.random_empty_book(
timesteps / 250, 1, 0.0, 0.05, 0.2, seed=69) # NOTE timesteps / 52 for DGA
# random_books.add_calls(init_instruments, book)
random_books.add_butterfly(init_instruments, book, 10)
# random_books.add_dga_calls(init_instruments, book)
# random_books.add_rko(init_instruments, book, 10.0)
driver = utils.HedgeDriver(timesteps=timesteps * hedge_multiplier,
frequency=frequency,
init_instruments=init_instruments,
init_numeraire=init_numeraire,
book=book,
cost=None,
risk_neutral=True,
learning_rate=1e-1)
# -*- coding: utf-8 -*-
import tensorflow as tf
import random_books
import utils
import hedge_models
import derivatives
# ==============================================================================
# === hyperparameters
train_size, test_size, timesteps = int(2**10), int(2**20), 30
alpha = 0.95
# ==============================================================================
# === sample data
init_instruments, init_numeraire, book = random_books.random_empty_book(
1.0, 1, 0.05, 0.1, 0.2, seed=69)
derivative = derivatives.PutCall(book.maturity, 105.0,
book.numeraire_simulator.rate,
book.instrument_simulator.volatility, 1)
book.add_derivative(derivative, 0, 1.0)
driver = utils.HedgeDriver(
timesteps=timesteps,
frequency=0, # no need for frequency
init_instruments=init_instruments,
init_numeraire=init_numeraire,
book=book,
cost=None,
risk_neutral=False,
learning_rate=1e-1)