\ud000\ud2b8 \ud2b8\ub808\uc774\ub529\uc740 \ub370\uc774\ud130 \uae30\ubc18\uc758 \uc804\ub7b5\uc744 \ud1b5\ud574 \uc790\ub3d9\uc73c\ub85c \ub9e4\ub9e4 \uacb0\uc815\uc744 \ub0b4\ub9ac\ub294 \uae30\ubc95\uc73c\ub85c, \uba38\uc2e0\ub7ec\ub2dd(ML) \ubc0f \ub525\ub7ec\ub2dd(DL) \uc54c\uace0\ub9ac\uc998\uc744 \ud65c\uc6a9\ud558\uc5ec \uc608\uce21 \ubaa8\ub378\uc744 \uac1c\ubc1c\ud558\ub294 \uac83\uc774 \ud575\uc2ec\uc785\ub2c8\ub2e4. \uc774 \uae00\uc5d0\uc11c\ub294 \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744 \uc0ac\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc758 \uc6d0\ub9ac, \ub2e4\uc591\ud55c \uc751\uc6a9 \uc0ac\ub840 \ubc0f \uc2e4\uc81c \uad6c\ud604 \ubc29\ubc95\uc5d0 \ub300\ud574 \uc54c\uc544\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n
\uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc740 \uc778\uacf5\uc9c0\ub2a5(AI) \ubd84\uc57c\uc758 \uc911\uc694\ud55c \ud558\uc704 \ubd84\uc57c\uc785\ub2c8\ub2e4. \uba38\uc2e0\ub7ec\ub2dd\uc740 \ub370\uc774\ud130\ub85c\ubd80\ud130 \ud559\uc2b5\ud574 \ud328\ud134\uc744 \uc778\uc2dd\ud558\uace0 \uc608\uce21\ud558\ub294 \uc54c\uace0\ub9ac\uc998\uc758 \uc9d1\ud569\uccb4\uc785\ub2c8\ub2e4. \ubc18\uba74 \ub525\ub7ec\ub2dd\uc740 \uc2e0\uacbd\ub9dd\uc744 \uae30\ubc18\uc73c\ub85c \ud55c \uba38\uc2e0\ub7ec\ub2dd\uc758 \ud55c \uc885\ub958\ub85c, \ud2b9\ud788 \ub300\uaddc\ubaa8 \ub370\uc774\ud130\uc5d0\uc11c \ubcf5\uc7a1\ud55c \ud328\ud134\uc744 \uc778\uc2dd\ud558\ub294 \ub370 \uac15\uc810\uc744 \uc9c0\ub2d9\ub2c8\ub2e4.<\/p>\n
\ub525\ub7ec\ub2dd\uc740 \uc5ec\ub7ec \uac1c\uc758 \uce35\uc73c\ub85c \uad6c\uc131\ub41c \uc2e0\uacbd\ub9dd(structured neural network)\uc744 \uc0ac\uc6a9\ud558\uc5ec \ub370\uc774\ud130\uc758 \ud2b9\uc9d5\uc744 \uc790\ub3d9\uc73c\ub85c \ucd94\ucd9c\ud569\ub2c8\ub2e4. \uac01 \uce35\uc740 \ub2e8\uc21c\ud558\uc9c0\ub9cc \uc870\ud569\ud558\uc5ec \ubcf5\uc7a1\ud55c \ubb38\uc81c\ub97c \ud574\uacb0\ud560 \uc218 \uc788\ub294 \ud2b9\uc9d5\uc744 \uac00\uc9d1\ub2c8\ub2e4.<\/p>\n
\uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \uc5ec\ub7ec \uad6c\uc131\uc694\uc18c\ub85c \uc774\ub8e8\uc5b4\uc838 \uc788\uc73c\uba70, \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998\uc740 \uc774 \uad6c\uc131\uc694\uc18c\ub97c \ucd5c\uc801\ud654\ud558\ub294 \ub370 \uc0ac\uc6a9\ub429\ub2c8\ub2e4.<\/p>\n
\ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc758 \uc131\uacf5\uc740 \ub370\uc774\ud130\uc758 \ud488\uc9c8\uc5d0 \ub2ec\ub824 \uc788\uc2b5\ub2c8\ub2e4. \uac00\uaca9 \ub370\uc774\ud130, \uac70\ub798\ub7c9 \ub370\uc774\ud130, \ub274\uc2a4 \ub370\uc774\ud130, \uc18c\uc15c \ubbf8\ub514\uc5b4 \ud53c\ub4dc \ub4f1 \ub2e4\uc591\ud55c \uc815\ubcf4\ub97c \uc218\uc9d1\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n
\uc218\uc9d1\ud55c \ub370\uc774\ud130\ub97c \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc5d0 \uc785\ub825\ud558\uae30 \uc804\uc5d0 \uc804\ucc98\ub9ac\uac00 \ud544\uc694\ud569\ub2c8\ub2e4. \uacb0\uce21\uce58 \ucc98\ub9ac, \uc815\uaddc\ud654, \uc6d0-\ud56b \uc778\ucf54\ub529 \ub4f1\uc758 \uacfc\uc815\uc774 \ud3ec\ud568\ub429\ub2c8\ub2e4.<\/p>\n
\ube44\uc988\ub2c8\uc2a4 \ubaa9\ud45c\uc5d0 \ub530\ub77c \uc801\ud569\ud55c \uba38\uc2e0\ub7ec\ub2dd \ub610\ub294 \ub525\ub7ec\ub2dd \ubaa8\ub378\uc744 \uc120\ud0dd\ud558\uace0 \ud559\uc2b5\uc2dc\ud0b5\ub2c8\ub2e4. \ub300\ud45c\uc801\uc778 \ubaa8\ub378\ub85c\ub294 \ud68c\uadc0 \ubd84\uc11d, \uc758\uc0ac \uacb0\uc815 \ud2b8\ub9ac, \ub79c\ub364 \ud3ec\ub808\uc2a4\ud2b8, LSTM(Long Short-Term Memory) \ub4f1\uc774 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n
\ubaa8\ub378\uc744 \ud1b5\ud574 \uac00\uaca9\uc774\ub098 \ud2b8\ub80c\ub4dc\uc5d0 \ub300\ud55c \uc608\uce21\uc744 \uc218\ud589\ud55c \ud6c4, \uc774\ub97c \uacfc\uac70 \ub370\uc774\ud130\uc5d0 \uc801\uc6a9\ud558\uc5ec \ubc31\ud14c\uc2a4\ud2b8\ub97c \uc9c4\ud589\ud574 \uc131\ub2a5\uc744 \ud3c9\uac00\ud569\ub2c8\ub2e4.<\/p>\n
\ubaa8\ub4e0 \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc5d0\ub294 \ub9ac\uc2a4\ud06c \uad00\ub9ac \uc804\ub7b5\uc774 \ud544\uc218\uc801\uc785\ub2c8\ub2e4. \uc190\uc2e4 \uc81c\ud55c, \ud3ec\uc9c0\uc158 \ud06c\uae30 \uc870\uc808 \ub4f1\uc744 \ud1b5\ud574 \uc190\uc2e4 \ub9ac\uc2a4\ud06c\ub97c \ucd5c\uc18c\ud654\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n
\uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc740 \uc5ec\ub7ec \uac00\uc9c0 \ubc29\uc2dd\uc73c\ub85c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc5d0 \ud65c\uc6a9\ub420 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \ub2e4\uc74c\uc740 \ub300\ud45c\uc801\uc778 \uc751\uc6a9 \uc0ac\ub840\uc785\ub2c8\ub2e4.<\/p>\n
\uc8fc\uac00 \uc608\uce21\uc740 \uba38\uc2e0\ub7ec\ub2dd\uc758 \ub300\ud45c\uc801\uc778 \uc751\uc6a9 \uc911 \ud558\ub098\uc785\ub2c8\ub2e4. \uc5ec\ub7ec \uc694\uc778(\uacfc\uac70 \uac00\uaca9, \uac70\ub798\ub7c9, \uacbd\uc81c \uc9c0\ud45c \ub4f1)\uc744 \ud1b5\ud574 \uc8fc\uac00\ub97c \uc608\uce21\ud558\ub294 \ubaa8\ub378\uc744 \uad6c\ucd95\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, LSTM \ub124\ud2b8\uc6cc\ud06c\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc2dc\uac04\uc758 \ud750\ub984\uc5d0 \ub530\ub77c \uc8fc\uac00 \ub370\uc774\ud130\ub97c \ud559\uc2b5\ud558\uace0 \uc608\uce21\ud558\ub294 \ubc29\ubc95\uc774 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n
\ud30c\uc774\uc36c LSTM \uc608\uc81c \ucf54\ub4dc<\/strong><\/p>\n \ud2b9\uc815 \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc744 \uad6c\ud604\ud560 \ub54c, \uba38\uc2e0\ub7ec\ub2dd \uae30\ubc95\uc744 \ud1b5\ud574 \ucd5c\uc801\uc758 \uc9c4\uc785 \ubc0f \uccad\uc0b0 \uc2e0\ud638\ub97c \ud3ec\ucc29\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4 \uc790\uc0b0 \uac04\uc758 \uc0c1\uad00 \uad00\uacc4, \uc774\ub3d9 \ud3c9\uade0 \uad50\ucc28 \uc804\ub7b5 \ub4f1\uc744 \uba38\uc2e0\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998\uc744 \ud1b5\ud574 \uc790\ub3d9\ud654\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n \ud30c\uc774\uc36c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc608\uc81c \ucf54\ub4dc<\/strong><\/p>\n \uc18c\uc15c \ubbf8\ub514\uc5b4, \ub274\uc2a4 \uae30\uc0ac\ub97c \ud1b5\ud574 \uc2dc\uc7a5 \uac10\uc815\uc744 \ubd84\uc11d\ud558\ub294 \uac83\ub3c4 \uac00\ub2a5\ud558\uba70, \uc774\ub97c \ud1b5\ud574 \uac00\uaca9 \ubcc0\ub3d9 \uc608\uce21\uc5d0 \ub3c4\uc6c0\uc774 \ub429\ub2c8\ub2e4. \uc790\uc5f0\uc5b4 \ucc98\ub9ac(NLP) \uae30\ubc95\uc744 \ud65c\uc6a9\ud558\uc5ec \ud14d\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \ubd84\uc11d\ud558\uace0 \uac10\uc815\uc744 \uc218\uce58\ud654\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc744 \ud1b5\ud574 \uac01 \uc790\uc0b0\uc758 \uc218\uc775\ub960 \ubc0f \ub9ac\uc2a4\ud06c\ub97c \uc608\uce21\ud558\uace0, \uc774\ub97c \ubc14\ud0d5\uc73c\ub85c \ud6a8\uc728\uc801\uc778 \ud3ec\ud2b8\ud3f4\ub9ac\uc624 \uad6c\uc131\uc744 \uc81c\uc548\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc5e1\uc2a4\ud0c0\uc778(Markowitz) \ud3ec\ud2b8\ud3f4\ub9ac\uc624 \uc774\ub860\uc744 \ud55c\uce35 \ubc1c\uc804\uc2dc\ud0a8 \uc5f0\uad6c\ub4e4\uc740 \ub354\uc6b1 \uc815\uad50\ud55c \uc790\uc0b0\ubc30\ubd84 \uc804\ub7b5\uc744 \uac00\ub2a5\ud558\uac8c \ud569\ub2c8\ub2e4.<\/p>\n \uc790\ub3d9 \ub9e4\ub9e4 \uc2dc\uc2a4\ud15c\uc740 \ub9ce\uc740 \uc7a0\uc7ac\uc801 \ub9ac\uc2a4\ud06c\ub97c \ub3d9\ubc18\ud569\ub2c8\ub2e4. \ub530\ub77c\uc11c, \uc2dc\uc2a4\ud15c\uc744 \ubc30\ud3ec\ud558\uae30 \uc804\uc5d0 \ucda9\ubd84\ud55c \ubc31\ud14c\uc2a4\ud2b8\uc640 \uac80\uc99d\uc744 \ud1b5\ud574 \uc2e0\ub8b0\uc131\uc744 \ud655\ubcf4\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc774 \ub108\ubb34 \ubcf5\uc7a1\ud560 \uacbd\uc6b0, \ud559\uc2b5 \ub370\uc774\ud130\uc5d0\ub294 \uc798 \ub9de\uc9c0\ub9cc \uc0c8\ub85c\uc6b4 \ub370\uc774\ud130\uc5d0\ub294 \uc798 \ub9de\uc9c0 \uc54a\uc744 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub97c \ubc29\uc9c0\ud558\uae30 \uc704\ud574 \ubaa8\ub378\uc758 \ub2e8\uc21c\ud654\ub97c \uace0\ub824\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n \ubc31\ud14c\uc2a4\ud2b8 \uacfc\uc815\uc5d0\uc11c \ubbf8\ub798 \uc815\ubcf4\ub97c \uc0ac\uc6a9\ud558\uac8c \ub418\ub294 \uacbd\uc6b0 \ub370\uc774\ud130 \uc2a4\ub108\ud0b9\uc774 \ubc1c\uc0dd\ud560 \uc218 \uc788\uc73c\ubbc0\ub85c, \uc774 \ub610\ud55c \uc8fc\uc758\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n \ub9ac\uc2a4\ud06c \uad00\ub9ac \uc804\ub7b5\uc744 \ubc18\ub4dc\uc2dc \ud3ec\ud568\uc2dc\ucf1c\uc57c \ud558\uba70, \uc774\uc775\uc744 \uadf9\ub300\ud654\ud558\uace0 \uc190\uc2e4\uc744 \ucd5c\uc18c\ud654\ud558\ub294 \ubc29\uc548\uc774 \ud544\uc694\ud569\ub2c8\ub2e4.<\/p>\n \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uae30\ubc95\uc740 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc5d0 \uc788\uc5b4 \uac15\ub825\ud55c \ub3c4\uad6c\uc774\uba70, \uc774\ub97c \ud65c\uc6a9\ud558\uc5ec \ub354 \ub098\uc740 \uc608\uce21\uacfc \uc804\ub7b5 \uac1c\ubc1c\uc774 \uac00\ub2a5\ud569\ub2c8\ub2e4. \ud558\uc9c0\ub9cc, \uc774\uc640 \ud568\uaed8 \ub9ac\uc2a4\ud06c \uad00\ub9ac \ubc0f \ucda9\ubd84\ud55c \ub370\uc774\ud130 \ubd84\uc11d\uc774 \uc120\ud589\ub418\uc5b4\uc57c \ud568\uc744 \uc78a\uc9c0 \ub9d0\uc544\uc57c \ud569\ub2c8\ub2e4. \uc2dc\uc7a5\uc740 \uacc4\uc18d\ud574\uc11c \ubcc0\ud654\ud558\uae30 \ub54c\ubb38\uc5d0 \uc9c0\uc18d\uc801\uc778 \ud559\uc2b5\uacfc \uac1c\uc120\uc744 \ud1b5\ud574 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c\uc744 \ubc1c\uc804\uc2dc\ucf1c \ub098\uac00\uc57c \ud569\ub2c8\ub2e4.<\/p>\n \uc774 \uae00\uc774 \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc5d0 \ub300\ud55c \uc774\ud574\ub97c \ub3d5\ub294 \ub370 \ub3c4\uc6c0\uc774 \ub418\uc5c8\uae30\ub97c \ubc14\ub78d\ub2c8\ub2e4.<\/p>\n <\/body><\/p>\n","protected":false},"excerpt":{"rendered":"\nimport numpy as np\nimport pandas as pd\nfrom keras.models import Sequential\nfrom keras.layers import LSTM, Dense, Dropout\n\n# \ub370\uc774\ud130 \ub85c\ub529 \ubc0f \uc804\ucc98\ub9ac\ndata = pd.read_csv('stock_data.csv')\ndata = data['Close'].values\n\n# \ub370\uc774\ud130\uc14b \uc0dd\uc131\ndef create_dataset(data, time_step=1):\n X, Y = [], []\n for i in range(len(data) - time_step - 1):\n a = data[i:(i + time_step)]\n X.append(a)\n Y.append(data[i + time_step])\n return np.array(X), np.array(Y)\n\nX, y = create_dataset(data, time_step=10)\nX = X.reshape(X.shape[0], X.shape[1], 1)\n\n# \ubaa8\ub378 \uad6c\ucd95\nmodel = Sequential()\nmodel.add(LSTM(50, return_sequences=True, input_shape=(X.shape[1], 1)))\nmodel.add(Dropout(0.2))\nmodel.add(LSTM(50, return_sequences=False))\nmodel.add(Dropout(0.2))\nmodel.add(Dense(1))\nmodel.compile(optimizer='adam', loss='mean_squared_error')\n\n# \ubaa8\ub378 \ud559\uc2b5\nmodel.fit(X, y, epochs=50, batch_size=32)\n <\/pre>\n<\/div>\n3.2 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc804\ub7b5 \uac1c\ubc1c<\/h3>\n
\nimport numpy as np\n\ndef moving_average(prices, window_size):\n return prices.rolling(window=window_size).mean()\n\ndef generate_signals(df):\n df['short_mavg'] = moving_average(df['Close'], window_size=10)\n df['long_mavg'] = moving_average(df['Close'], window_size=30)\n \n # \ub9e4\uc218 \uc2e0\ud638\n df['signal'] = 0\n df.loc[df['short_mavg'] > df['long_mavg'], 'signal'] = 1\n df.loc[df['short_mavg'] <= df['long_mavg'], 'signal'] = -1\n\n return df\n\n# \uc608\uc2dc \ub370\uc774\ud130\ud504\ub808\uc784 \uc0dd\uc131\ndf = pd.DataFrame({'Close': [100, 101, 102, 100, 99, 98, 99, 100, 101, 102]})\ndf = generate_signals(df)\n <\/pre>\n<\/div>\n3.3 \uc2dc\uc7a5 \uac10\uc815 \ubd84\uc11d<\/h3>\n
3.4 \ud3ec\ud2b8\ud3f4\ub9ac\uc624 \ucd5c\uc801\ud654<\/h3>\n
4. \uae30\ud0c0 \uace0\ub824\uc0ac\ud56d<\/h2>\n
4.1 \uacfc\uc801\ud569(Overfitting) \ubb38\uc81c<\/h3>\n
4.2 \ub370\uc774\ud130 \uc2a4\ub108\ud0b9(Data Snooping)<\/h3>\n
4.3 \ub9ac\uc2a4\ud06c \uad00\ub9ac<\/h3>\n
5. \uacb0\ub860<\/h2>\n