Arima实战：利用Python中pyramid-arima库进行时间序列预测

Arima 模型的重要假设

ARMA, ARIMA, SARIMA assumptions:
▪ Time-series data is stationary.
▪ If nonstationary, remove trend, seasonality, apply differencing, and so on.
▪ Remember that stationary data has no trend, seasonality, constant mean, and
constant variance.
▪ Therefore, the past is assumed to represent what will happen in the future
in a probabilistic sense.

1. 安装

pyramid-arima 的安装请见 https://pypi.org/project/pyramid-arima/
(我只在linux系统上成功安装了，windows上没有成功)

$ pip install pyramid-arima

函数参数介绍请见 https://www.alkaline-ml.com/pmdarima/index.html

Github上的例子请见 https://github.com/tgsmith61591/pmdarima

2. 代码实例

import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
import pmdarima as pm
import warnings
warnings.filterwarnings("ignore")

2.1 加载数据集

df = pd.read_csv('dataset.csv')

split_point = 1000
data_train = df['x'].iloc[:split_point].values
data_test = df['x'].iloc[split_point:].values

2.2 训练模型

pm.auto_arima可以自动搜索出arima模型中的(q, d, p)参数

p--代表预测模型中采用的时序数据本身的滞后数(lags) ,也叫做AR/Auto-Regressive项
d--代表时序数据需要进行几阶差分化，才是稳定的，也叫Integrated项
q--代表预测模型中采用的预测误差的滞后数(lags)，也叫做MA/Moving Average项

参考 https://blog.csdn.net/HHXUN/article/details/79858672

model = pm.auto_arima(data_train)

2.3 模型预测

利用 model.predict() 函数预测

x_pred = model.predict(n_periods=1)

或更优的，使用 model.update() 函数，不断用新观测到的 value 更新模型，以达到更长时间的预测。

pred_list = []
for x_i in data_test:
    pred_list += [model.predict(n_periods=1)]
    # 更新模型
    model.update(x_i)

2.4 模型评价

from sklearn import metrics
def eval_metrics(y_true, y_pred):
    metrics_dict = dict()
    metrics_dict['MAE'] = metrics.mean_absolute_error(y_true, y_pred)
    metrics_dict['MSE'] = metrics.mean_squared_error(y_true, y_pred)
    metrics_dict['MAPE'] = np.mean(np.true_divide(np.abs(y_true-y_pred), y_true))
    return metrics_dict

eval_dict = eval_metrics(data_test,  np.array(pred_list))
print(eval_dict)

作者：菜鸟程序猿zq
链接：https://www.jianshu.com/p/54826718be4f
来源：简书
著作权归作者所有。商业转载请联系作者获得授权，非商业转载请注明出处。