arima-LSTM混合模型代码

以下是一个简单的Python代码，演示如何使用ARIMA和LSTM进行时间序列预测的混合模型：

import pandas as pd
import numpy as np
from statsmodels.tsa.arima_model import ARIMA
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM

# 载入数据
data = pd.read_csv('data.csv', header=None, index_col=0, parse_dates=True, squeeze=True)

# 拆分数据集为训练集和测试集
train_size = int(len(data) * 0.8)
train, test = data[0:train_size], data[train_size:len(data)]

# ARIMA模型
model_arima = ARIMA(train, order=(5, 1, 0))
model_arima_fit = model_arima.fit(disp=0)
print(model_arima_fit.summary())

# LSTM模型
def create_dataset(dataset, look_back=1):
    dataX, dataY = [], []
    for i in range(len(dataset)-look_back-1):
        a = dataset[i:(i+look_back), 0]
        dataX.append(a)
        dataY.append(dataset[i + look_back, 0])
    return np.array(dataX), np.array(dataY)

look_back = 3
trainX, trainY = create_dataset(train.values, look_back)
testX, testY = create_dataset(test.values, look_back)

trainX = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))
testX = np.reshape(testX, (testX.shape[0], 1, testX.shape[1]))

model_lstm = Sequential()
model_lstm.add(LSTM(4, input_shape=(1, look_back)))
model_lstm.add(Dense(1))
model_lstm.compile(loss='mean_squared_error', optimizer='adam')
model_lstm.fit(trainX, trainY, epochs=100, batch_size=1, verbose=2)

# 模型预测
train_predict_arima = model_arima_fit.predict(start=1, end=len(train), dynamic=False)
test_predict_arima = model_arima_fit.predict(start=len(train)+1, end=len(data), dynamic=False)

train_predict_lstm = model_lstm.predict(trainX)
test_predict_lstm = model_lstm.predict(testX)

# 将ARIMA和LSTM模型的预测结果合并
train_predict = (train_predict_arima + train_predict_lstm.reshape(train_predict_arima.shape)) / 2
test_predict = (test_predict_arima + test_predict_lstm.reshape(test_predict_arima.shape)) / 2

# 输出预测结果
print(train_predict)
print(test_predict)

注：此代码仅供参考，具体实现需要根据实际数据进行调整和优化。