[Python] Assign column status retrospectively in pandas

I have created the following pandas dataframe:

import pandas as pd
import numpy as np

ds = {'col1' : [234,321,284,286,287,300,301,303,305,299,288,300,299,287,286,280,279,270,269,301]}

df = pd.DataFrame(data=ds)


The dataframe looks like this:

display(df)

col1
0 234
1 321
2 284
3 286
4 287
5 300
6 301
7 303
8 305
9 299
10 288
11 300
12 299
13 287
14 286
15 280
16 279
17 270
18 269
19 301


I have then created two columns (cnsIncr , cnsDecr), which calculate the consecutive increase and decrease in col1 respectively.

cnsIncr = []
cnsDecr = []

col1 = np.array(df['col1'])

for i in range(len(df)):
cnsIncr.append(0)
cnsDecr.append(0)
if(col1 > col1[i-1]):
cnsIncr = cnsIncr[i-1]+1
else:
cnsIncr = 0

if(col1 < col1[i-1]):
cnsDecr = cnsDecr[i-1]+1
else:
cnsDecr = 0

df['cnsIncr'] = cnsIncr
df['cnsDecr'] = cnsDecr


The resulting dataframe looks as follows:



I need to create a columns called Trend which is populated as follows:

• if the column cnsIncr contains a value greater or equal to 5, then Trend = UpTrend starting from the record in which consIncr = 0 and ending at the record for which cnsIncr >= 5;
• if the column cnsDecr contains a value greater or equal to 5, then Trend = DownTrend starting from the record in which cnsDecr = 0 and ending at the record for which cnsDecr >= 5;
• the neither the column cnsIncr nor the column cnsDecr contain a value of 5, then Trend = NoTrend

So, from the example above, the resulting dataframe would look like this:



How can I do this?

Continue reading...