Question

我有一个如下所示的数据框：

Station       A           B          C
Date
2013-01-31  1340381     1568766     910785
2013-02-28  1261806     1447467     843956
2013-03-31  1399123     1579597     926968  
2013-04-30  1395016     1618159     950947 
2013-05-31  1340408     1654265     988293

每个月每个车站的总人数。我如何计算2013年哪一个电台增长最快（旅客人数增加）？

Answer 1

您可以进行线性回归并估算一年内的增长率。当然，由于您没有一年的数据，因此您会对季节性变化产生偏见。

import numpy as np
import pandas as pd
from scipy import stats

df =  pd.DataFrame()

df['A'] = np.linspace(1000, 2000, 6)
df['B'] = np.linspace(1000, 3000, 6)
df['C'] = np.linspace(1000, 4000, 6)
df.index = pd.date_range('1/1/2015', periods=6, freq='M', name='Date')

print df

t = (pd.to_datetime(df.index.values) - pd.to_datetime('2015-01-01')).astype('timedelta64[D]')

slope, intercept, r_value, p_value, std_err = stats.linregress(t, df['A'])
print ('\nEstimated growth in one year of line A: ' + str(slope*365))

slope, intercept, r_value, p_value, std_err = stats.linregress(t, df['B'])
print ('\nEstimated growth in one year of line B: ' + str(slope*365))

slope, intercept, r_value, p_value, std_err = stats.linregress(t, df['C'])
print ('\nEstimated growth in one year of line C: ' + str(slope*365))

根据月度数据计算年度增长率

1 个答案: