我正在寻找使用可在此处下载的天然气季节性数据构建 ARMA 模型。
所以:
import statsmodels.api as sm
import pandas as pd
df = pd.read_csv('/Path/To/WSJ-NG_HH.csv')
In [86]: df.head
Out[86]:
<bound method DataFrame.head of Date Value
0 2015-10-31 2.100
1 2015-09-30 2.470
2 2015-08-31 2.680
3 2015-07-31 2.770
4 2015-06-30 2.770
接下来我np.asarray()
在 DataFrame 上应用 data =:
data = np.asarray(df)
然后实例化ARMA
:
arma = sm.tsa.ARMA(data, order =(4,4))
当我尝试适应时:
results = arma.fit(full_output=False, disp=0)
我得到:
/Users/Pyderman/anaconda/lib/python2.7/site-packages/statsmodels/regression/linear_model.pyc in fit(self, method, cov_type, cov_kwds, use_t, **kwargs)
180 self.rank = np_matrix_rank(np.diag(singular_values))
181
--> 182 beta = np.dot(self.pinv_wexog, self.wendog)
183
184 elif method == "qr":
TypeError: can't multiply sequence by non-int of type 'float'
必须做什么才能处理数据中的浮点值?
统计模型版本:0.6.1