Consider below dataFrame. I want to calculate if the current value of price column is greater than last 10 values. I was thinking to use shift, but not sure how to use for last 10 rows.

    price
220 3.337
221 3.320
222 3.290
223 3.291
224 3.312
225 3.255
226 3.216
227 3.245
228 3.275
229 3.282
230 3.370
231 3.396
232 3.375
233 3.369
234 3.335
235 3.344
236 3.365
237 3.373
238 3.414
239 3.378

Output dataframe:

    price  isGreater
220 3.337      NaN
221 3.320      NaN
222 3.290      NaN
223 3.291      NaN
224 3.312      NaN
225 3.255      NaN
226 3.216      NaN
227 3.245      NaN
228 3.275      NaN
229 3.282      NaN
230 3.370      1.0
231 3.396      1.0
232 3.375      NaN
233 3.369      NaN
234 3.335      NaN
235 3.344      NaN
236 3.365      NaN
237 3.373      NaN
238 3.414      1.0
239 3.378      NaN

CodePudding user response：

You can use rolling max to get the max of the last 10 rows, if greater than it, then it's greater or equal than all (including self, thus the 1):

df['isGreater'] = df['price'].ge(df['price'].rolling(10 1).max())

NB. technically if you really want to compare only to the previous rows and not self (for example to use strict comparison), you would need to shift:

df['isGreater'] = df['price'].gt(df['price'].shift().rolling(10).max())

output:

     price  isGreater
220  3.337      False
221  3.320      False
222  3.290      False
223  3.291      False
224  3.312      False
225  3.255      False
226  3.216      False
227  3.245      False
228  3.275      False
229  3.282      False
230  3.370       True
231  3.396       True
232  3.375      False
233  3.369      False
234  3.335      False
235  3.344      False
236  3.365      False
237  3.373      False
238  3.414       True
239  3.378      False