├── DQN_Pair.ipynb
└── README.md
/DQN_Pair.ipynb:
--------------------------------------------------------------------------------
1 | {
2 | "nbformat": 4,
3 | "nbformat_minor": 0,
4 | "metadata": {
5 | "colab": {
6 | "name": "DQN_Pair.ipynb",
7 | "provenance": [],
8 | "authorship_tag": "ABX9TyOy9bvsV0isIHhfWvaABFV0",
9 | "include_colab_link": true
10 | },
11 | "kernelspec": {
12 | "name": "python3",
13 | "display_name": "Python 3"
14 | },
15 | "language_info": {
16 | "name": "python"
17 | }
18 | },
19 | "cells": [
20 | {
21 | "cell_type": "markdown",
22 | "metadata": {
23 | "id": "view-in-github",
24 | "colab_type": "text"
25 | },
26 | "source": [
27 | "
"
28 | ]
29 | },
30 | {
31 | "cell_type": "code",
32 | "metadata": {
33 | "colab": {
34 | "base_uri": "https://localhost:8080/"
35 | },
36 | "id": "pidTUe2PiU50",
37 | "outputId": "a23383bf-cda3-4486-b04a-a93fefe575d5"
38 | },
39 | "source": [
40 | "from google.colab import drive\n",
41 | "drive.mount('/content/drive')"
42 | ],
43 | "execution_count": 1,
44 | "outputs": [
45 | {
46 | "output_type": "stream",
47 | "name": "stdout",
48 | "text": [
49 | "Mounted at /content/drive\n"
50 | ]
51 | }
52 | ]
53 | },
54 | {
55 | "cell_type": "code",
56 | "metadata": {
57 | "id": "tQEM9z5tiluy"
58 | },
59 | "source": [
60 | "import numpy as np\n",
61 | "import random\n",
62 | "import math\n",
63 | "import gym\n",
64 | "import sys\n",
65 | "import pandas as pd\n",
66 | "import matplotlib.pyplot as plt\n",
67 | "import datetime\n",
68 | "from datetime import datetime, timedelta\n",
69 | "import csv\n",
70 | "from collections import deque\n",
71 | "\n",
72 | "import keras\n",
73 | "from keras.models import Sequential\n",
74 | "from keras.models import load_model\n",
75 | "from keras.layers import Dense\n",
76 | "from tensorflow.keras.optimizers import Adam"
77 | ],
78 | "execution_count": 2,
79 | "outputs": []
80 | },
81 | {
82 | "cell_type": "code",
83 | "metadata": {
84 | "id": "8XsToq63iscy"
85 | },
86 | "source": [
87 | "def formatPrice(n): #Print fomatted price\n",
88 | " return (\"-$\" if n < 0 else \"$\") + \"{:0.2f}\".format(abs(n))\n",
89 | "\n",
90 | "def getStockDataVec(key): #returns stock data vector\n",
91 | " vec = []\n",
92 | " lines = open(\"data/\" + key + \".txt\", \"r\").read().splitlines()\n",
93 | " for line in lines[1:]:\n",
94 | " vec.append(float(line.split(\",\")[4]))\n",
95 | "\n",
96 | " return vec\n",
97 | "\n",
98 | "def getStockVolVec(key): #returns stock volume vector\n",
99 | " vol = []\n",
100 | " lines = open(\"/content/drive/MyDrive/data/crypto_market_data_candlesticks\" + key + \".txt\", \"r\").read().splitlines()\n",
101 | " for line in lines[1:]:\n",
102 | " vol.append(float(line.split(\",\")[5]))\n",
103 | "\n",
104 | " return vol\n",
105 | "\n",
106 | "def sigmoid(x):\n",
107 | " return 1 / (1 + math.exp(-x))"
108 | ],
109 | "execution_count": 3,
110 | "outputs": []
111 | },
112 | {
113 | "cell_type": "code",
114 | "metadata": {
115 | "id": "6MYL7uo0jtw8"
116 | },
117 | "source": [
118 | "import numpy as np\n",
119 | "import random\n",
120 | "import math, random \n",
121 | "import gym \n",
122 | "import numpy as np \n",
123 | "\n",
124 | "\n",
125 | "class State:\n",
126 | " def __init__(self, data1, data2, Bal_stock1, Bal_stock2, open_cash, timestep):\n",
127 | " self.Stock1Price=data1[timestep] #stock 1 open price\n",
128 | " self.Stock2Price=data2[timestep] #stock 2 open price\n",
129 | " self.Stock1Blnc=Bal_stock1 #stock 1 balance\n",
130 | " self.Stock2Blnc=Bal_stock2 #stock 2 balance\n",
131 | " self.open_cash=open_cash #cash balance\n",
132 | " self.fiveday_stock1=self.five_day_window(data1, timestep)\n",
133 | " self.fiveday_stock2=self.five_day_window(data2, timestep)\n",
134 | " #self.volume1=volume1[timestep]\n",
135 | " #self.volume2=volume2[timestep]\n",
136 | " self.portfolio_value=self.portfolio_value()\n",
137 | "\n",
138 | " def portfolio_value(self):\n",
139 | " pvalue=0\n",
140 | " #print(\"In portfolio func\")\n",
141 | " #print(\"self.Stock1Price\",self.Stock1Price, type(self.Stock1Price))\n",
142 | " #print(\"self.Stock1Blnc\",self.Stock1Blnc[0], type(self.Stock1Blnc))\n",
143 | "\n",
144 | " v1=self.Stock1Price * float(self.Stock1Blnc)\n",
145 | " v2=self.Stock2Price * float(self.Stock2Blnc)\n",
146 | " v3=(float(self.open_cash)/2)\n",
147 | " return (v1+v2+v3)\n",
148 | " \n",
149 | " def next_opening_price(self):\n",
150 | " return [data1[timestep+1], data2[timestep+1]]\n",
151 | " \n",
152 | " def five_day_window(self,data, timestep):\n",
153 | " step = timestep\n",
154 | " if step < 5:\n",
155 | " return data[0]\n",
156 | " \n",
157 | " stock_5days = np.mean(data[step-5:step])\n",
158 | " #print(\"stock_5days=\" + str(stock_5days))\n",
159 | " #print(stock_5days)\n",
160 | "\n",
161 | " #print(type(stock_5days))\n",
162 | "\n",
163 | " return stock_5days\n",
164 | " \n",
165 | " def reset(self, data1, data2, Bal_stock1, Bal_stock2, open_cash, timestep):\n",
166 | " #self.state = torch.FloatTensor(torch.zeros(8)).cuda()\n",
167 | " self.Stock1Price=data1[timestep] #stock 1 open price \n",
168 | " self.Stock2Price=data2[timestep] #stock 2 open price \n",
169 | " self.Stock1Blnc=Bal_stock1 #stock 1 balance \n",
170 | " self.Stock2Blnc=Bal_stock2 #stock 2 balance \n",
171 | " self.open_cash=open_cash #cash balance\n",
172 | " self.fiveday_stock1=self.five_day_window(data1, timestep)\n",
173 | " self.fiveday_stock2=self.five_day_window(data2, timestep)\n",
174 | " self.portfolio_value=10000\n",
175 | " \n",
176 | " def getState(self):\n",
177 | " #print(\"In get state\")\n",
178 | " res=[]\n",
179 | " res.append(self.Stock1Price) #stock 1 open price\n",
180 | " res.append(self.Stock2Price) #stock 2 open price\n",
181 | " res.append(self.Stock1Blnc) #stock 1 balance\n",
182 | " res.append(self.Stock2Blnc) #stock 2 balance\n",
183 | " res.append(self.open_cash) #cash balance\n",
184 | " res.append(self.fiveday_stock1)\n",
185 | " res.append(self.fiveday_stock2) \n",
186 | " res.append(self.portfolio_value)\n",
187 | " #res.append(self.volume1)\n",
188 | " #res.append(self.volume2)\n",
189 | "\n",
190 | "\n",
191 | " \n",
192 | " #print(res)\n",
193 | " res1=np.array([res])\n",
194 | " #print(\"res array\"+np.array([res]))\n",
195 | " return res1"
196 | ],
197 | "execution_count": 4,
198 | "outputs": []
199 | },
200 | {
201 | "cell_type": "code",
202 | "metadata": {
203 | "id": "9R-qn-Zdjv-T"
204 | },
205 | "source": [
206 | "class Agent:\n",
207 | " def __init__(self, state_size, is_eval=False, model_name=\"\"):\n",
208 | " self.state_size = state_size\n",
209 | " self.action_size = 5 #buy1, sell1, buy2, sell2, do nothing.\n",
210 | " self.memory = deque(maxlen=2000)\n",
211 | " self.inventory1 = []\n",
212 | " self.inventory2 = []\n",
213 | " self.model_name = model_name\n",
214 | " self.is_eval = is_eval\n",
215 | " self.gamma = 0.95 #discount factor (quantifies how much importance to give future rewards)\n",
216 | " self.epsilon = 1.0 #Eploration and Exploitation\n",
217 | " self.epsilon_min = 0.01\n",
218 | " self.epsilon_decay = 0.995\n",
219 | " self.model = load_model(\"/content/drive/MyDrive/IDS576/Project/models/\" + model_name) if is_eval else self._model()\n",
220 | " \n",
221 | " def _model(self):\n",
222 | " model = Sequential()\n",
223 | " model.add(Dense(units=64, input_dim=self.state_size, activation='relu'))\n",
224 | " model.add(Dense(units=32, activation='relu'))\n",
225 | " model.add(Dense(units=8, activation='relu'))\n",
226 | " model.add(Dense(self.action_size, activation='linear'))\n",
227 | " model.compile(loss=\"mse\", optimizer=Adam(lr=0.0001))\n",
228 | " return model\n",
229 | " \n",
230 | " def act(self, state):\n",
231 | " if not self.is_eval and random.random() <= self.epsilon:\n",
232 | " return random.randrange(self.action_size)\n",
233 | " options = self.model.predict(state)\n",
234 | " return np.argmax(options[0])\n",
235 | "\n",
236 | " def expReplay(self, batch_size):\n",
237 | " mini_batch = []\n",
238 | " l = len(self.memory)\n",
239 | "\n",
240 | " minibatch = random.sample(self.memory, batch_size)\n",
241 | "\n",
242 | " for state, action, reward, next_state, done in mini_batch:\n",
243 | " target = reward\n",
244 | "\n",
245 | " if not done:\n",
246 | " target = reward + self.gamma * np.amax(self.model.predict(next_state)[0])\n",
247 | " \n",
248 | " target_f = self.model.predict(state)\n",
249 | " target_f[0][action] = target\n",
250 | " self.model.fit(state, target_f, epochs=1, verbose=0)\n",
251 | " \n",
252 | " if self.epsilon > self.epsilon_min:\n",
253 | " self.epsilon *= self.epsilon_decay"
254 | ],
255 | "execution_count": 5,
256 | "outputs": []
257 | },
258 | {
259 | "cell_type": "code",
260 | "metadata": {
261 | "id": "6vLHaYZJj0U_"
262 | },
263 | "source": [
264 | "stock_name1, stock_name2, episode_count, start_balance, training, test = 'aapl.us', 'amzn.us', 51, 10000, 1500, 500"
265 | ],
266 | "execution_count": 6,
267 | "outputs": []
268 | },
269 | {
270 | "cell_type": "code",
271 | "metadata": {
272 | "id": "mNWZkjFbj2I9"
273 | },
274 | "source": [
275 | "pd_data1 = pd.read_csv('/content/drive/MyDrive/data/crypto_market_data_candlesticks/BTCUSDT10.csv', sep=\",\", header=0)\n",
276 | "pd_data2 = pd.read_csv('/content/drive/MyDrive/data/crypto_market_data_candlesticks/ETHUSDT10.csv', sep=\",\", header=0)"
277 | ],
278 | "execution_count": 7,
279 | "outputs": []
280 | },
281 | {
282 | "cell_type": "code",
283 | "metadata": {
284 | "id": "w2S76bS0kBIa"
285 | },
286 | "source": [
287 | "pd_data1['Date']=pd.to_datetime(pd_data1['Date'], format='%Y/%m/%d')\n",
288 | "pd_data2['Date']=pd.to_datetime(pd_data2['Date'], format='%Y/%m/%d')"
289 | ],
290 | "execution_count": 8,
291 | "outputs": []
292 | },
293 | {
294 | "cell_type": "code",
295 | "metadata": {
296 | "colab": {
297 | "base_uri": "https://localhost:8080/"
298 | },
299 | "id": "_8l-Xgt8kGUl",
300 | "outputId": "b1584cc6-6b09-49b6-9f9b-f51823c5f782"
301 | },
302 | "source": [
303 | "pd_data2['Date'][0]"
304 | ],
305 | "execution_count": 9,
306 | "outputs": [
307 | {
308 | "output_type": "execute_result",
309 | "data": {
310 | "text/plain": [
311 | "Timestamp('2020-01-01 00:00:00')"
312 | ]
313 | },
314 | "metadata": {},
315 | "execution_count": 9
316 | }
317 | ]
318 | },
319 | {
320 | "cell_type": "code",
321 | "metadata": {
322 | "colab": {
323 | "base_uri": "https://localhost:8080/"
324 | },
325 | "id": "6ocK4auDkJwq",
326 | "outputId": "94676a07-9cd9-4271-ef52-039bf8c2facd"
327 | },
328 | "source": [
329 | "if (pd_data1['Date'][0] > pd_data2['Date'][0]):\n",
330 | " print(\"pd_data1 is older\")\n",
331 | " pd_data1 = pd_data1[pd_data1.Date>=pd_data2['Date'][0]]\n",
332 | " pd_data1 = pd_data1.reset_index(drop=True)\n",
333 | "else:\n",
334 | " print(\"pd_data1 is not older\")\n",
335 | " pd_data2 = pd_data2[pd_data2.Date>=pd_data1['Date'][0]]\n",
336 | " pd_data2 = pd_data2.reset_index(drop=True)"
337 | ],
338 | "execution_count": 10,
339 | "outputs": [
340 | {
341 | "output_type": "stream",
342 | "name": "stdout",
343 | "text": [
344 | "pd_data1 is not older\n"
345 | ]
346 | }
347 | ]
348 | },
349 | {
350 | "cell_type": "code",
351 | "metadata": {
352 | "id": "1-VzkOHHkL7F"
353 | },
354 | "source": [
355 | "import datetime\n",
356 | "list1 = pd_data1['Date']\n",
357 | "list2 = pd_data2['Date']\n",
358 | "diff_pd1_data = list(set(list1) - set(list2))\n",
359 | "diff_pd2_data = list(set(list2) - set(list1))\n",
360 | "for k in range(len(diff_pd1_data)):\n",
361 | " pd1_date_format = diff_pd1_data[k].strftime('%Y-%m-%d 00:00:00')\n",
362 | " date_format_pd1 = datetime.datetime.strptime(pd1_date_format, '%Y-%m-%d 00:00:00')\n",
363 | " for i,j in enumerate(list1):\n",
364 | " if j == date_format_pd1:\n",
365 | " pd_data1 = pd_data1.drop([i])\n",
366 | "pd_data1 = pd_data1.reset_index(drop=True)\n",
367 | "\n",
368 | "for k in range(len(diff_pd2_data)):\n",
369 | " pd2_date_format = diff_pd2_data[k].strftime('%Y-%m-%d 00:00:00')\n",
370 | " date_format_pd2 = datetime.datetime.strptime(pd2_date_format, '%Y-%m-%d 00:00:00')\n",
371 | " for l,m in enumerate(list2):\n",
372 | " if m == date_format_pd2:\n",
373 | " pd_data2 = pd_data2.drop([l])\n",
374 | "pd_data2 = pd_data2.reset_index(drop=True)"
375 | ],
376 | "execution_count": 11,
377 | "outputs": []
378 | },
379 | {
380 | "cell_type": "code",
381 | "metadata": {
382 | "colab": {
383 | "base_uri": "https://localhost:8080/",
384 | "height": 295
385 | },
386 | "id": "WH55voBpkPL7",
387 | "outputId": "ac3ec38e-463f-4aa6-a05e-9c8a03a8f37a"
388 | },
389 | "source": [
390 | "%matplotlib inline\n",
391 | "\n",
392 | "x1 = np.array(pd_data1['Date'])\n",
393 | "y1 = pd_data1['Open']\n",
394 | "y12 = pd_data1['Volume']\n",
395 | "\n",
396 | "plt.title(\"BTC\")\n",
397 | "plt.xlabel('Year')\n",
398 | "plt.ylabel(\"Price in $\")\n",
399 | "\n",
400 | "plt.plot(x1,y1)\n",
401 | "\n",
402 | "ax2 = plt.twinx()\n",
403 | "\n",
404 | "color = 'tab:red'\n",
405 | "ax2.set_ylabel('volume', color=color)\n",
406 | "ax2.plot(x1, y12, color=color)\n",
407 | "ax2.tick_params(axis='y', labelcolor=color)"
408 | ],
409 | "execution_count": 12,
410 | "outputs": [
411 | {
412 | "output_type": "display_data",
413 | "data": {
414 | "image/png": "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\n",
415 | "text/plain": [
416 | ""
417 | ]
418 | },
419 | "metadata": {
420 | "needs_background": "light"
421 | }
422 | }
423 | ]
424 | },
425 | {
426 | "cell_type": "code",
427 | "metadata": {
428 | "colab": {
429 | "base_uri": "https://localhost:8080/",
430 | "height": 295
431 | },
432 | "id": "FMdn75ujkUk7",
433 | "outputId": "92c4d00c-a74d-46b7-ef3f-b7311521d506"
434 | },
435 | "source": [
436 | "%matplotlib inline\n",
437 | "\n",
438 | "x2 = np.array(pd_data2['Date'])\n",
439 | "y2 = pd_data2['Open']\n",
440 | "y22 = pd_data1['Volume']\n",
441 | "\n",
442 | "plt.title(\"ETH\")\n",
443 | "plt.xlabel('Year')\n",
444 | "plt.ylabel(\"Price in $\")\n",
445 | "\n",
446 | "plt.plot(x2,y2)\n",
447 | "\n",
448 | "ax2 = plt.twinx()\n",
449 | "\n",
450 | "color = 'tab:red'\n",
451 | "ax2.set_ylabel('volume', color=color)\n",
452 | "ax2.plot(x2, y22, color=color)\n",
453 | "ax2.tick_params(axis='y', labelcolor=color)\n",
454 | "plt.show()"
455 | ],
456 | "execution_count": 13,
457 | "outputs": [
458 | {
459 | "output_type": "display_data",
460 | "data": {
461 | "image/png": "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\n",
462 | "text/plain": [
463 | ""
464 | ]
465 | },
466 | "metadata": {
467 | "needs_background": "light"
468 | }
469 | }
470 | ]
471 | },
472 | {
473 | "cell_type": "code",
474 | "metadata": {
475 | "id": "o4tm41zjkYs0"
476 | },
477 | "source": [
478 | "pd_data1_train = pd_data1[0:training]\n",
479 | "pd_data2_train = pd_data2[0:training]\n",
480 | "pd_data1_test = pd_data1[training:training+test]\n",
481 | "pd_data2_test = pd_data2[training:training+test]"
482 | ],
483 | "execution_count": 14,
484 | "outputs": []
485 | },
486 | {
487 | "cell_type": "code",
488 | "metadata": {
489 | "id": "GpxOfr90kb_R"
490 | },
491 | "source": [
492 | ""
493 | ],
494 | "execution_count": null,
495 | "outputs": []
496 | }
497 | ]
498 | }
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # DQN_pairtrading
2 | Optimizing the Pairs-Trading Strategy using Deep Reinforcement Learning with Trading and Stop-loss Boundaries
3 |
--------------------------------------------------------------------------------