def connect(): logger = logging.getLogger('client') logger.setLevel('ERROR') print("Setup was successful.") e = Exchange() if not e.is_connected(): e.connect() print('connected') else: print('already connected') return e
import time from optibook.synchronous_client import Exchange import logging logger = logging.getLogger('client') logger.setLevel('ERROR') print("Setup was successful.") # compare the bid and ask prices of the two instruments e = Exchange() a = e.connect() instrument_limit = 190 instrument_limit1 = 100 orderBatch = 0 def get_info(): #print("in get_info") print(e.get_positions()) positions = e.get_positions() for p in positions: print(p, positions[p]) pnl = e.get_pnl() print("current pnl is", pnl)
class AutoTrader: """ This is the "main" class which houses our algorithm. You will see there are a few helper functions already here, as well as a main "trade" function which runs the algorithm. We've done some work for you already there, but you will need to write the bulk of the strategy yourself. """ def __init__(self): self.exchange_client = Exchange() def connect(self): """ Connect to the optibook exchange """ self.exchange_client.connect() def get_order_book_for_instrument(self, instrument): return self.exchange_client.get_last_price_book(instrument) def get_position_for_instrument(self, instrument): positions = self.exchange_client.get_positions() return positions[instrument] def get_top_of_book(self, order_book): """ Get the best bid and best ask of the order book you pass in as a parameter. """ best_bid_price = None best_bid_volume = None if len(order_book.bids) > 0: best_bid_price = round(order_book.bids[0].price, 2) best_bid_volume = round(order_book.bids[0].volume, 2) best_ask_price = None best_ask_volume = None if len(order_book.asks) > 0: best_ask_price = round(order_book.asks[0].price, 2) best_ask_volume = round(order_book.asks[0].volume, 2) return TopOfBook(best_bid_price, best_bid_volume, best_ask_price, best_ask_volume) def print_top_of_book(self, instrument, top_of_book): print( f'[{instrument}] bid({top_of_book.best_bid_volume}@{top_of_book.best_bid_price})-ask({top_of_book.best_ask_volume}@{top_of_book.best_ask_price})' ) def insert_buy_order(self, instrument, price, volume, order_type): """ Insert an order to buy. Note that volume must be positive. Also note that you have no guarantee that your order turns into a trade. instrument: str The name of the instrument to buy. price: float The price level at which to insert the order into the order book on the bid side. volume: int The volume to buy. order_type: int You can set this to 'limit' or 'ioc'. 'limit' orders stay in the book while any remaining volume of an 'ioc' that is not immediately matched is cancelled. return: an InsertOrderReply containing a request_id as well as an order_id, the order_id can be used to e.g. delete or amend the limit order later. """ return self.exchange_client.insert_order(instrument, price=price, volume=volume, side='bid', order_type=order_type) def insert_sell_order(self, instrument, price, volume, order_type): """ Insert an order to sell. Note that volume must be positive. Also note that you have no guarantee that your order turns into a trade. instrument: str The name of the instrument to sell. price: float The price level at which to insert the order into the order book on the ask side. volume: int The volume to sell. order_type: int You can set this to 'limit' or 'ioc'. 'limit' orders stay in the book while any remaining volume of an 'ioc' that is not immediately matched is cancelled. return: an InsertOrderReply containing a request_id as well as an order_id, the order_id can be used to e.g. delete or amend the limit order later. """ return self.exchange_client.insert_order(instrument, price=price, volume=volume, side='ask', order_type=order_type) def increase(self, array): last = array[0] for element in array: if (element < last): return False last = element return True def decrease(self, array): last = array[0] for element in array: if (element > last): return False last = element return True def trade(self): """ This function is the main trading algorithm. It is called in a loop, and in every iteration of the loop we do the exact same thing. We start by getting the order books, formatting them a little bit and then you will have to make a trading decision based on the prices in the order books. """ # First we get the current order books of both instruments full_book_liquid = self.get_order_book_for_instrument( LIQUID_INSTRUMENT) full_book_illiquid = self.get_order_book_for_instrument( ILLIQUID_INSTRUMENT) # Then we extract the best bid and best ask from those order books top_book_liquid = self.get_top_of_book(full_book_liquid) top_book_illiquid = self.get_top_of_book(full_book_illiquid) # If either the bid side or ask side is missing, in the order books, then we stop right here and wait for the # next cycle, in the hopes that then the order books will have both the bid and ask sides present if not top_book_liquid.has_bid_and_ask( ) or not top_book_illiquid.has_bid_and_ask(): print( 'There are either no bids or no asks, skipping this trade cycle.' ) return # Print the top of each book, this will be very helpful to you when you want to understand what your # algorithm is doing. Feel free to add more logging as you see fit. self.print_top_of_book(LIQUID_INSTRUMENT, top_book_liquid) self.print_top_of_book(ILLIQUID_INSTRUMENT, top_book_illiquid) print('') # Trade! # Take if from here, and implement your actual strategy with the help of the pre-processing we have done for you # above. Note that this is very rudimentary, and there are things we have left out (e.g. position management is # missing, hedging is missing, and how much credit you ask for is also missing). # # Maybe a first step is to run this code as is, and see what it prints out to get some inspiration if you are # stuck. Otherwise, come to us, we are always happy to help. Check the client documentation for all the # functions that are at your disposal. # # ----------------------------------------- # TODO: Implement trade logic here ''' instruments = ['PHILIPS_A', 'PHILIPS_B'] SIZE = 5 for index, instrument in enumerate(instruments): asks[index].append(self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_ask_price) if len(asks[index]) > SIZE: asks[index].pop(0) bids[index].append(self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_bid_price) if len(bids[index]) > SIZE: bids[index].pop(0) print(asks) print(bids) positions = self.exchange_client.get_positions() stocks = positions[instrument] if (len(bids[index]) == SIZE and self.increase(bids[index])): print("stocks"+str(stocks)) doTrade = self.insert_sell_order(instrument, self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_bid_price, max(1, int(stocks * 1/5)), 'ioc') # come back and change volume print("sell") elif len(asks[index]) == SIZE and self.decrease(asks[index]) and (positions[instruments[1]] + positions[instruments[0]]) <200: print("positions: " + str(positions[instrument])) doTrade = self.insert_buy_order(instrument, self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_ask_price, 1, 'ioc') #change volvume print("buy") ''' ''' instruments = ['PHILIPS_A', 'PHILIPS_B'] SIZE = 100 for index, instrument in enumerate(instruments): aux = self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_ask_price if aux: asks[index].append(aux) if len(asks[index]) > SIZE: asks[index].pop(0) aux = self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_bid_price bids[index].append(aux) if len(bids[index]) > SIZE: bids[index].pop(0) if SIZE == len(asks[index]): averageAsk = sum(asks[index]) / len(asks) averageBid = sum(bids[index]) / len(bids) positions = self.exchange_client.get_positions() stocks = positions[instrument] if averageBid > self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_bid_price and abs(stocks) < 100: print(abs(stocks)) doTrade = self.insert_buy_order(instrument, self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_bid_price - 0.5, , 'ioc') print("buy") if averageAsk < self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_ask_price : doTrade = self.insert_sell_order(instrument, self.get_top_of_book(self.get_order_book_for_instrument(instrument)).best_ask_price + 0.5, max(1, int(stocks * 1/2)), 'ioc') # come back and change volume print("sell") ''' bidA = None askB = None instruments = ['PHILIPS_A', 'PHILIPS_B'] SIZE = 1 while (not askB) or (not bidA): bidA = self.get_top_of_book( self.get_order_book_for_instrument( instruments[0])).best_bid_price askB = self.get_top_of_book( self.get_order_book_for_instrument( instruments[1])).best_ask_price if bidA - askB > 0 and askB < 1000: doTrade = self.insert_sell_order(instruments[0], bidA, SIZE, 'ioc') doTrade = self.insert_buy_order(instruments[1], askB, SIZE, 'ioc') print("bidA and askB") bidB = None askA = None while (not askA) or (not bidB): bidB = self.get_top_of_book( self.get_order_book_for_instrument( instruments[1])).best_bid_price askA = self.get_top_of_book( self.get_order_book_for_instrument( instruments[0])).best_ask_price if bidB - askA > 0 and askA < 1000: doTrade = self.insert_sell_order(instruments[1], bidB, SIZE, 'ioc') doTrade = self.insert_buy_order(instruments[0], askA, SIZE, 'ioc') print("bidB and askA")
class AutoTrader: """ This is the "main" class which houses our algorithm. You will see there are a few helper functions already here, as well as a main "trade" function which runs the algorithm. We've done some work for you already there, but you will need to write the bulk of the strategy yourself. """ def __init__(self): self.exchange_client = Exchange() def connect(self): """ Connect to the optibook exchange """ self.exchange_client.connect() def get_order_book_for_instrument(self, instrument): return self.exchange_client.get_last_price_book(instrument) def get_position_for_instrument(self, instrument): positions = self.exchange_client.get_positions() return positions[instrument] def get_top_of_book(self, order_book): """ Get the best bid and best ask of the order book you pass in as a parameter. """ best_bid_price = None best_bid_volume = None if len(order_book.bids) > 0: best_bid_price = round(order_book.bids[0].price, 2) best_bid_volume = round(order_book.bids[0].volume, 2) best_ask_price = None best_ask_volume = None if len(order_book.asks) > 0: best_ask_price = round(order_book.asks[0].price, 2) best_ask_volume = round(order_book.asks[0].volume, 2) return TopOfBook(best_bid_price, best_bid_volume, best_ask_price, best_ask_volume) def print_top_of_book(self, instrument, top_of_book): print( f'[{instrument}] bid({top_of_book.best_bid_volume}@{top_of_book.best_bid_price})-ask({top_of_book.best_ask_volume}@{top_of_book.best_ask_price})' ) def insert_buy_order(self, instrument, price, volume, order_type): """ Insert an order to buy. Note that volume must be positive. Also note that you have no guarantee that your order turns into a trade. instrument: str The name of the instrument to buy. price: float The price level at which to insert the order into the order book on the bid side. volume: int The volume to buy. order_type: int You can set this to 'limit' or 'ioc'. 'limit' orders stay in the book while any remaining volume of an 'ioc' that is not immediately matched is cancelled. return: an InsertOrderReply containing a request_id as well as an order_id, the order_id can be used to e.g. delete or amend the limit order later. """ return self.exchange_client.insert_order(instrument, price=price, volume=volume, side='bid', order_type=order_type) def insert_sell_order(self, instrument, price, volume, order_type): """ Insert an order to sell. Note that volume must be positive. Also note that you have no guarantee that your order turns into a trade. instrument: str The name of the instrument to sell. price: float The price level at which to insert the order into the order book on the ask side. volume: int The volume to sell. order_type: int You can set this to 'limit' or 'ioc'. 'limit' orders stay in the book while any remaining volume of an 'ioc' that is not immediately matched is cancelled. return: an InsertOrderReply containing a request_id as well as an order_id, the order_id can be used to e.g. delete or amend the limit order later. """ return self.exchange_client.insert_order(instrument, price=price, volume=volume, side='ask', order_type=order_type) def trade(self): """ This function is the main trading algorithm. It is called in a loop, and in every iteration of the loop we do the exact same thing. We start by getting the order books, formatting them a little bit and then you will have to make a trading decision based on the prices in the order books. """ # First we get the current order books of both instruments full_book_liquid = self.get_order_book_for_instrument( LIQUID_INSTRUMENT) full_book_illiquid = self.get_order_book_for_instrument( ILLIQUID_INSTRUMENT) # Then we extract the best bid and best ask from those order books top_book_liquid = self.get_top_of_book(full_book_liquid) top_book_illiquid = self.get_top_of_book(full_book_illiquid) # If either the bid side or ask side is missing, in the order books, then we stop right here and wait for the # next cycle, in the hopes that then the order books will have both the bid and ask sides present if not top_book_liquid.has_bid_and_ask( ) or not top_book_illiquid.has_bid_and_ask(): print( 'There are either no bids or no asks, skipping this trade cycle.' ) return # Print the top of each book, this will be very helpful to you when you want to understand what your # algorithm is doing. Feel free to add more logging as you see fit. self.print_top_of_book(LIQUID_INSTRUMENT, top_book_liquid) self.print_top_of_book(ILLIQUID_INSTRUMENT, top_book_illiquid) print('')
class Bot: instruments = ["PHILIPS_A", "PHILIPS_B"] def __init__(self): self.e = Exchange() logging.info(self.e.connect()) logging.info("Setup was successful.") def get_out_of_positions(self): # Get out of all positions you are currently holding, regardless of the loss involved. That means selling whatever # you are long, and buying-back whatever you are short. Be sure you know what you are doing when you use this logic. print(self.e.get_positions()) for s, p in self.e.get_positions().items(): if p > 0: self.e.insert_order(s, price=1, volume=p, side='ask', order_type='ioc') elif p < 0: self.e.insert_order(s, price=100000, volume=-p, side='bid', order_type='ioc') print(self.e.get_positions()) # Logging functions def log_new_trade_ticks(self): logger.info("Polling new trade ticks") for i in self.instruments: tradeticks = self.e.poll_new_trade_ticks(i) for t in tradeticks: logger.info( f"[{t.instrument_id}] price({t.price}), volume({t.volume}), aggressor_side({t.aggressor_side}), buyer({t.buyer}), seller({t.seller})" ) def log_positions_cash(self): logger.info(self.e.get_positions_and_cash()) def log_all_outstanding_orders(self): for i in self.instruments: logger.info(self.e.get_outstanding_orders(i)) def wait_until_orders_complete(self): orders_outstanding = True while orders_outstanding: orders_outstanding = False for i in self.instruments: if len(self.e.get_outstanding_orders(i)) > 0: orders_outstanding = True self.log_all_outstanding_orders() #time.sleep(0.1) def mainloop(self): while True: # check for trade differences # m1 ask < m2 bid #logger.info("Checking for discrepancies:") books = [self.e.get_last_price_book(x) for x in self.instruments] for m1, m2 in [(0, 1), (1, 0)]: m1_id = self.instruments[m1] m2_id = self.instruments[m2] try: m1_ask = books[m1].asks[0] m2_bid = books[m2].bids[0] if m1_ask.price < m2_bid.price: logger.info( f"Can profit: buy {m1_id} at {m1_ask} and sell {m2_id} at {m2_bid}" ) self.e.insert_order(m1_id, price=m1_ask.price, volume=1, side='bid', order_type='limit') self.e.insert_order(m2_id, price=m2_bid.price, volume=1, side='ask', order_type='limit') self.log_all_outstanding_orders() self.wait_until_orders_complete() self.log_positions_cash() except Exception as e: print(logger.error(e)) continue time.sleep(1.0 / 25)
import time import logging from optibook.synchronous_client import Exchange from strategy import should_kill_attempt, arbitrage, stoikov_mm from utils import balance_positions from moving_average import MovingAverage logging.getLogger('client').setLevel('ERROR') exchange = Exchange() exchange.connect() START_PNL = exchange.get_pnl() ma_A = MovingAverage(exchange, "PHILIPS_A") tick = 1 while not should_kill_attempt(exchange, START_PNL): time.sleep(0.11) print(f"tick {tick}") tick += 1 ma_A.update() # Don't want to balance our trades from our MM positions exchange.delete_orders("PHILIPS_A")
def connect(): global exchange exchange = Exchange() return exchange.connect()
import safe_dual_strat import difference_strat import extremes_strat import dual_paper_strat import dual_hedge_strat import mimic_strat import sys logger = logging.getLogger('client') logger.setLevel('INFO') # Main loop: run a number of different strategies if __name__ == '__main__': exchange = Exchange() logging.info(exchange.connect()) logging.info("Setup was successful.") try: # we can run multiple strategies at once, but really having separate bots to do this would be simpler # since currently we can't ignore our own orders in aggregate data such as the pricebook # strats = [dual_market_strat.DualMarketStrat(exchange, ["PHILIPS_A", "PHILIPS_B"])] # strats = [safe_dual_strat.SafeDualStrat(exchange, ["PHILIPS_A", "PHILIPS_B"])] # strats = [difference_strat.DifferenceStrat(exchange, ["PHILIPS_A", "PHILIPS_B"])] #strats = [dual_paper_strat.DualPaperStrat(exchange, ["PHILIPS_A", "PHILIPS_B"])] strats = [dual_hedge_strat.DualHedgeStrat(exchange, "PHILIPS_A", "PHILIPS_B")] while True: time.sleep(0.2) for s in strats: try: s.update(); except AssertionError as e: # if get disconnected than do something else