CODEWITHSUNDEEP
pythonpandas
David Glasser
David Glasser

Reputation: 31

Percent change from one column to the next

Let's say I had a dataframe:

df = pd.DataFrame([[3, 2, 1], [5, 4, 2]])

3 2 1
5 4 2

I want to return a dataframe that has the percent change from one column to the next. So the above dataset would return:

.666 .5
.8   .5

How would I accomplish this in pandas?

Upvotes: 0

Views: 185

Answers (2)

Ricky
Ricky

Reputation: 2750

Not a very ideal way but works.

df['diff_1_0']=abs(df[0]-df[1])/df[0]
df['diff_2_1']=abs(df[1]-df[2])/df[1]
df=df[['diff_1_0','diff_2_1']]

If you have a lot of columns you can make it into a loop like below.

column_count=df.shape[1]+1
column_names=df.columns
for columns in df:
    if columns==0:
        continue
    else:
        df.insert(columns, str('diff_'+str(column_count)), abs(df[columns-1]-df[columns])/df[columns-1])
        column_count=column_count+1
df=df.drop(column_names, axis=1)

Hope it helps!

Upvotes: 0

naccode
naccode

Reputation: 520

Pandas has a function for this. Use pct_change:

df.pct_change(axis='columns')

This will output the 'percentage change' (per your question) from one column to other. Not the division of one column to the other (seems to be your example).

Upvotes: 1

Related Questions