\n \ucd5c\uadfc \uba87 \ub144 \ub3d9\uc548 \uae08\uc735 \uc2dc\uc7a5\uc5d0\uc11c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc774 \uae09\uaca9\ud788 \ubc1c\uc804\ud588\uc2b5\ub2c8\ub2e4. \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd \uae30\uc220\uc740 \ub370\uc774\ud130 \ubd84\uc11d\uacfc \uc608\uce21 \uc815\ud655\ub3c4\ub97c \ud5a5\uc0c1\uc2dc\ud0a4\ub294\ub370 \ud070 \ub3c4\uc6c0\uc744 \uc8fc\uc5c8\uace0, \uc774\uc5d0 \ub530\ub77c \uc790\ub3d9\ub9e4\ub9e4 \uc2dc\uc2a4\ud15c\uc758 \uad6c\uc131\uacfc \uc6b4\uc601\uc5d0 \uc788\uc5b4 \ud544\uc218\uc801\uc778 \uc694\uc18c\ub85c \uc790\ub9ac \uc7a1\uc558\uc2b5\ub2c8\ub2e4. \ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd\uc744 \uae30\ubc18\uc73c\ub85c \ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc758 \uae30\ubcf8 \uac1c\ub150\uacfc, \ud2b9\uc131\uc744 \uc774\ud574\ud558\uace0, \uc774\ub97c \ud1b5\ud574 \uc2dc\uac04\uc5d0 \ub530\ub77c \ub370\uc774\ud130 \uc0dd\uc131 \ubc0f \ucc98\ub9ac\ub97c \ubc30\uc6b0\uaca0\uc2b5\ub2c8\ub2e4.\n <\/p>\n<\/div>\n
\n \uba38\uc2e0\ub7ec\ub2dd\uc740 \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ud328\ud134\uc744 \ud559\uc2b5\ud558\uace0 \uc608\uce21\ud558\ub294 \uc54c\uace0\ub9ac\uc998\uc785\ub2c8\ub2e4. \uae30\uacc4 \ud559\uc2b5\uc758 \uc8fc\uc694 \ubaa9\ud45c\ub294 \uc8fc\uc5b4\uc9c4 \uc785\ub825 \ub370\uc774\ud130\uc5d0\uc11c \uacb0\uacfc\ub97c \ub3c4\ucd9c\ud574\ub0b4\ub294 \uac83\uc785\ub2c8\ub2e4. \ub525\ub7ec\ub2dd\uc740 \uba38\uc2e0\ub7ec\ub2dd\uc758 \ud55c \ubd84\uc57c\ub85c, \uc778\uacf5\uc2e0\uacbd\ub9dd\uc744 \uc0ac\uc6a9\ud558\uc5ec \ub354\uc6b1 \ubcf5\uc7a1\ud55c \ud328\ud134\uc744 \ud559\uc2b5\ud569\ub2c8\ub2e4. \uc8fc\uc2dd\uc2dc\uc7a5\uc5d0\uc11c\ub294 \ub2e4\uc74c\uacfc \uac19\uc740 \uc138 \uac00\uc9c0 \uc720\ud615\uc758 \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc774 \uc790\uc8fc \uc0ac\uc6a9\ub429\ub2c8\ub2e4.\n <\/p>\n
\n \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc744 \uc704\ud574\uc11c\ub294 \uc6b0\uc120 \uae08\uc735 \ub370\uc774\ud130\uac00 \ud544\uc694\ud569\ub2c8\ub2e4. \uc8fc\ub85c \uc8fc\uc2dd \uac00\uaca9, \uac70\ub798\ub7c9, \uae30\uc220 \ubd84\uc11d \uc9c0\ud45c \ub4f1\uc758 \ub370\uc774\ud130\uac00 \uc0ac\uc6a9\ub429\ub2c8\ub2e4. \uc774\ub7ec\ud55c \ub370\uc774\ud130\ub294 \ub2e4\uc591\ud55c API\ub098 \uae08\uc735 \ub370\uc774\ud130 \uc81c\uacf5 \uc0ac\uc774\ud2b8\ub97c \ud1b5\ud574 \uc218\uc9d1\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, Python\uc758 \n \uc218\uc9d1\ud55c \ub370\uc774\ud130\ub294 \uc885\uc885 \ub178\uc774\uc988\ub098 \ubd88\uc644\uc804\uc131\uc744 \ud3ec\ud568\ud558\uace0 \uc788\uc2b5\ub2c8\ub2e4. \ub530\ub77c\uc11c \ubd84\uc11d\uc5d0 \uc801\ud569\ud55c \ud615\ud0dc\ub85c \uc804\ucc98\ub9ac\ud574\uc57c \ud569\ub2c8\ub2e4. \uc804\ucc98\ub9ac \ub2e8\uacc4\ub294 \uc77c\ubc18\uc801\uc73c\ub85c \uacb0\uce21\uce58 \ucc98\ub9ac, \uc815\uaddc\ud654, \uc2a4\ucf00\uc77c\ub9c1 \ub4f1\uc758 \uc791\uc5c5\uc744 \ud3ec\ud568\ud569\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, \ub370\uc774\ud130\uc758 \uc2a4\ucf00\uc77c\uc744 \ub9de\ucd94\uae30 \uc704\ud574 Min-Max \uc2a4\ucf00\uc77c\ub7ec\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc8fc\uac00 \ub370\uc774\ud130\ub97c 0\uacfc 1 \uc0ac\uc774\ub85c \ubcc0\ud658\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n \n \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc758 \uc131\ub2a5\uc740 \ud2b9\uc131(feature)\uc5d0 \ud06c\uac8c \uc758\uc874\ud569\ub2c8\ub2e4. \ub530\ub77c\uc11c \uc801\uc808\ud55c \ud2b9\uc131\uc744 \uc0dd\uc131\ud558\uace0 \uc120\ud0dd\ud558\ub294 \uacfc\uc815\uc774 \ub9e4\uc6b0 \uc911\uc694\ud569\ub2c8\ub2e4. \uc8fc\ub85c \uc0ac\uc6a9\ub418\ub294 \ud2b9\uc131 \uc0dd\uc131 \ubc29\ubc95\uc5d0\ub294 \uc774\ub3d9 \ud3c9\uade0, \uc0c1\ub300 \uac15\ub3c4 \uc9c0\uc218(RSI) \ub4f1\uacfc \uac19\uc740 \uae30\uc220\uc801 \ubd84\uc11d \uc9c0\ud45c \ubc0f \ud1b5\uacc4\uc801 \uc9c0\ud45c\ub4e4\uc774 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n \n \uc774\ub3d9 \ud3c9\uade0\uc740 \ud2b9\uc815 \uae30\uac04 \ub3d9\uc548\uc758 \ud3c9\uade0 \uac00\uaca9\uc744 \uacc4\uc0b0\ud558\uc5ec \uc8fc\uac00\uc758 \ucd94\uc138\ub97c \ud310\ub2e8\ud558\ub294 \ub370 \uc0ac\uc6a9\ub429\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, 20\uc77c \uc774\ub3d9 \ud3c9\uade0\uc744 \uacc4\uc0b0\ud558\ub294 \ucf54\ub4dc\ub294 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4.\n <\/p>\n \n RSI\ub294 \uc8fc\uac00\uac00 \uacfc\ub9e4\ub3c4 \ub610\ub294 \uacfc\ub9e4\uc218 \uc0c1\ud0dc\uc778\uc9c0 \ud310\ub2e8\ud558\ub294 \ub370 \uc0ac\uc6a9\ub418\ub294 \uc9c0\ud45c\uc785\ub2c8\ub2e4. RSI\ub97c \uacc4\uc0b0\ud558\ub824\uba74 \uc0c1\uc2b9\uacfc \ud558\ub77d\uc758 \ud3c9\uade0\uc744 \ud65c\uc6a9\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n \n \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\ub85c\ub294 \ud68c\uadc0 \ubaa8\ub378, \uc758\uc0ac\uacb0\uc815\ub098\ubb34, \ub79c\ub364 \ud3ec\ub808\uc2a4\ud2b8, SVM \ub4f1\uc774 \uc788\uc73c\uba70, \ub525\ub7ec\ub2dd \ubaa8\ub378\ub85c\ub294 LSTM(Long Short-Term Memory), CNN(Convolutional Neural Network) \ub4f1\uc774 \uc788\uc2b5\ub2c8\ub2e4. \uac01 \ubaa8\ub378\uc740 \uace0\uc720\uc758 \ud2b9\uc131\uacfc \uc7a5\ub2e8\uc810\uc774 \uc788\uc73c\uba70, \uc8fc\uc5b4\uc9c4 \ubb38\uc81c\uc5d0 \ub9de\ub294 \ubaa8\ub378\uc744 \uc120\ud0dd\ud558\ub294 \uac83\uc774 \uc911\uc694\ud569\ub2c8\ub2e4.\n <\/p>\n \n \uc120\ud0dd\ud55c \ubaa8\ub378\uc744 \ud6c8\ub828 \ub370\uc774\ud130\uc5d0 \ub9de\ucdb0 \ud559\uc2b5\uc2dc\ud0b5\ub2c8\ub2e4. \ubaa8\ub378 \ud559\uc2b5\uc740 \uc77c\ubc18\uc801\uc73c\ub85c \uc190\uc2e4 \ud568\uc218\ub97c \ucd5c\uc18c\ud654\ud558\ub294 \ubc29\ud5a5\uc73c\ub85c \uc9c4\ud589\ub429\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, TensorFlow\ub97c \uc0ac\uc6a9\ud558\uc5ec LSTM \ubaa8\ub378\uc744 \uad6c\uc131\ud558\ub294 \ucf54\ub4dc\ub294 \uc544\ub798\uc640 \uac19\uc2b5\ub2c8\ub2e4.\n <\/p>\n \n \ubaa8\ub378 \ud3c9\uac00 \ub2e8\uacc4\uc5d0\uc11c\ub294 \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \uc0ac\uc6a9\ud558\uc5ec \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \uce21\uc815\ud569\ub2c8\ub2e4. \uc5ec\ub7ec \uc9c0\ud45c, \uc608\ub97c \ub4e4\uc5b4 MSE(Mean Squared Error)\ub97c \uc0ac\uc6a9\ud558\uc5ec \ud3c9\uac00\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \ub610\ud55c, \ud559\uc2b5\ud55c \ubaa8\ub378\uc758 \uc608\uce21 \uacb0\uacfc\ub97c \uc2dc\uac01\ud654\ud558\uc5ec \uc9c1\uad00\uc801\uc73c\ub85c \ud3c9\uac00\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n \n \uc774\uc81c\uae4c\uc9c0 \ud559\uc2b5\ud55c \ub0b4\uc6a9\ub4e4\uc744 \ubc14\ud0d5\uc73c\ub85c \uc2e4\uc81c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c\uc744 \uad6c\ucd95\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774 \uc2dc\uc2a4\ud15c\uc740 \uc8fc\uc5b4\uc9c4 \ub370\uc774\ud130\ub85c\ubd80\ud130 \uacb0\uc815\uc744 \ub0b4\ub9ac\uace0 \uc790\ub3d9\uc73c\ub85c \uc8fc\ubb38\uc744 \uc2e4\ud589\ud558\ub294 \uae30\ub2a5\uc744 \ud3ec\ud568\ud558\uac8c \ub429\ub2c8\ub2e4.\n <\/p>\n \n \uac70\ub798 \uc2e0\ud638\ub294 \uc8fc\uc2dd\uc758 \ub9e4\uc218 \ub610\ub294 \ub9e4\ub3c4 \uc2dc\uc810\uc744 \uacb0\uc815\ud558\ub294 \uc9c0\ud45c\uc785\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, \uc774\ub3d9 \ud3c9\uade0 \uad50\ucc28\ubcf4\uae30\ub97c \uc774\uc6a9\ud55c \ub9e4\ub9e4 \uc2e0\ud638\ub97c \uc0dd\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc544\ub798\uc758 \ucf54\ub4dc\ub294 \ub2e8\uc21c\ud55c \ub9e4\ub9e4 \uc804\ub7b5\uc744 \uad6c\ud604\ud558\ub294 \uc608\uc785\ub2c8\ub2e4.\n <\/p>\n \n \uac70\ub798 \uc2e0\ud638\uac00 \uc0dd\uc131\ub418\uba74, \uc774\ub97c \ubc14\ud0d5\uc73c\ub85c \uc2e4\uc81c \uc8fc\ubb38\uc744 \uc2e4\ud589\ud569\ub2c8\ub2e4. \ub300\ubd80\ubd84\uc758 \ud2b8\ub808\uc774\ub529 \ud50c\ub7ab\ud3fc\uc740 API\ub97c \ud1b5\ud574 \uc790\ub3d9\uc73c\ub85c \uc8fc\ubb38\uc744 \uc2e4\ud589\ud560 \uc218 \uc788\ub3c4\ub85d \uc9c0\uc6d0\ud558\uba70, \ud3ec\ud2b8\ud3f4\ub9ac\uc624\uc758 \uc131\uacfc\ub97c \uad00\ub9ac\ud558\ub294 \uae30\ub2a5\ub3c4 \ud3ec\ud568\ub418\uc5b4\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n \n \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \uae08\uc735 \uc2dc\uc7a5\uc5d0\uc11c \uc218\uc775\uc744 \uc5bb\uae30 \uc704\ud55c \uac15\ub825\ud55c \ub3c4\uad6c\uc785\ub2c8\ub2e4. \ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 \ub370\uc774\ud130 \uc218\uc9d1, \uc804\ucc98\ub9ac, \ud2b9\uc131 \uc0dd\uc131, \ubaa8\ub378 \ud559\uc2b5\uacfc \ud3c9\uac00, \uc2e4\uc81c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uad6c\ucd95\uae4c\uc9c0\uc758 \uacfc\uc815\uc744 \ub2e4\ub8e8\uc5c8\uc2b5\ub2c8\ub2e4. \ubb34\uc5c7\ubcf4\ub2e4\ub3c4, \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc758 \uc131\uacf5\uc740 \uc9c0\uc18d\uc801\uc778 \ub370\uc774\ud130 \ubd84\uc11d\uacfc \ubaa8\ub378 \uac1c\uc120\uc5d0 \uc788\uc74c\uc744 \uae30\uc5b5\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n<\/div>\nyfinance<\/code> \ub77c\uc774\ube0c\ub7ec\ub9ac\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc8fc\uc2dd \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n\nimport yfinance as yf\n\n# \uc560\ud50c\uc758 \uc8fc\uc2dd \ub370\uc774\ud130\ub97c \uac00\uc838\uc635\ub2c8\ub2e4.\ndata = yf.download(\"AAPL\", start=\"2020-01-01\", end=\"2023-01-01\")\nprint(data.head())\n <\/pre>\n<\/div>\n
3.2. \ub370\uc774\ud130 \uc804\ucc98\ub9ac<\/h3>\n
\nfrom sklearn.preprocessing import MinMaxScaler\n\n# MinMaxScaler \ucd08\uae30\ud654 \ubc0f \uc801\ud569\nscaler = MinMaxScaler()\nscaled_data = scaler.fit_transform(data[['Close']])\n <\/pre>\n<\/div>\n<\/div>\n
4. \ud2b9\uc131 \uc0dd\uc131 \ubc0f \uc120\ud0dd<\/h2>\n
4.1. \uc774\ub3d9 \ud3c9\uade0(Moving Average)<\/h3>\n
\ndata['SMA_20'] = data['Close'].rolling(window=20).mean()\n <\/pre>\n<\/div>\n
4.2. \uc0c1\ub300 \uac15\ub3c4 \uc9c0\uc218(Relative Strength Index, RSI)<\/h3>\n
\ndef compute_rsi(data, window=14):\n delta = data['Close'].diff()\n gain = (delta.where(delta > 0, 0)).rolling(window=window).mean()\n loss = (-delta.where(delta < 0, 0)).rolling(window=window).mean()\n rs = gain \/ loss\n rsi = 100 - (100 \/ (1 + rs))\n return rsi\n\ndata['RSI'] = compute_rsi(data)\n <\/pre>\n<\/div>\n<\/div>\n
5. \ubaa8\ub378 \ud559\uc2b5 \ubc0f \ud3c9\uac00<\/h2>\n
5.1. \ubaa8\ub378 \uc120\ud0dd<\/h3>\n
5.2. \ubaa8\ub378 \ud559\uc2b5<\/h3>\n
\nimport tensorflow as tf\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import LSTM, Dense\n\nmodel = Sequential()\nmodel.add(LSTM(50, activation='relu', input_shape=(X_train.shape[1], X_train.shape[2])))\nmodel.add(Dense(1))\nmodel.compile(optimizer='adam', loss='mean_squared_error')\nmodel.fit(X_train, y_train, epochs=50, batch_size=32)\n <\/pre>\n<\/div>\n
5.3. \ubaa8\ub378 \ud3c9\uac00<\/h3>\n
\nimport matplotlib.pyplot as plt\n\npredictions = model.predict(X_test)\nplt.plot(y_test, label='\uc2e4\uc81c \uac12')\nplt.plot(predictions, label='\uc608\uce21 \uac12')\nplt.legend()\nplt.show()\n <\/pre>\n<\/div>\n<\/div>\n
6. \uc2e4\uc81c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uad6c\ucd95<\/h2>\n
6.1. \uac70\ub798 \uc2e0\ud638 \uc0dd\uc131<\/h3>\n
\ndata['Signal'] = 0\ndata['Signal'][20:] = np.where(data['SMA_20'][20:] > data['Close'][20:], 1, 0)\n <\/pre>\n<\/div>\n
6.2. \uc8fc\ubb38 \uc2e4\ud589 \ubc0f \ud3ec\ud2b8\ud3f4\ub9ac\uc624 \uad00\ub9ac<\/h3>\n
\nimport requests\n\ndef send_order(signal):\n if signal == 1:\n # \ub9e4\uc218 \uc8fc\ubb38 \uc2e4\ud589 \ucf54\ub4dc\n requests.post(\"API_ENDPOINT\", data={\"action\": \"buy\", \"quantity\": 1})\n elif signal == -1:\n # \ub9e4\ub3c4 \uc8fc\ubb38 \uc2e4\ud589 \ucf54\ub4dc\n requests.post(\"API_ENDPOINT\", data={\"action\": \"sell\", \"quantity\": 1})\n <\/pre>\n<\/div>\n<\/div>\n7. \uacb0\ub860<\/h2>\n