Question

我正在做一个数据科学项目，试图利用以前的警察站点来预测未来的站点。我正在尝试绘制时间与地理空间经度的自相关图，但是当我尝试绘制该图时，我总是遇到错误。这使我无法找到建立ARIMA模型的正确滞后。

from pandas import datetime
from pandas import to_datetime
from matplotlib import pyplot
from pandas.plotting import autocorrelation_plot

dataSet = read_excel('SeniorProject.xlsx', header=0,
                   encoding = "ISO-8859-1")
mini = dataSet[['TIME_PHONEPICKUP','GEO_LON']]
#print(mini)

#mini.plot(x='TIME_PHONEPICKUP',y='GEO_LON', color='blue')
autocorrelation_plot(mini)

pyplot.show()

错误

TypeError: unsupported operand type(s) for +: 'Timestamp' and 'Timestamp'

UserWarning: Requested projection is different from current axis projection, creating new axis with requested projection.
  ax = plt.gca(xlim=(1, n), ylim=(-1.0, 1.0))
During handling of the above exception, another exception occurred:
KeyError: '__name__'

一些样本数据

   TIME_PHONEPICKUP     GEO_LON
0 2019-09-09 03:57:08 -105.025060
1 2019-09-09 03:55:39 -104.990248
2 2019-09-09 03:52:32 -104.925776
3 2019-09-09 03:49:44 -105.032151
4 2019-09-09 03:45:23 -105.029842

TIME_PHONEPICKUP是dtype：datetime64 [ns]，而GEO_LON是dtype：float64。我是否正确设置了自相关图？

Answer 1

我认为问题出在时间结构上，也许您应该尝试编制时间索引，看看如何 ACP去了。

也许是这样的：

autocorrelation_plot(df.set_index('time'))
plt.show()

请参阅熊猫pandas dataframe set_index

的文档

创建具有不规则时间序列的自相关图？

1 个答案: