Find stock with highest variance and swap it with first column of dataframe-CodePudding

I have a dataframe containing m series of stock prices for different companies like the one below:

Date  Stock 1   Stock 2  Stock 3 ...  Stock m
1      100       200      300    ...   500
2      500       300      200    ...   100
:       :         :        :     ...    :
n      200       300      400    ...   100

What I need to do is find out which of the m stocks has the highest variance and then swap it with the entries of Stock 1. Assume that in the above example Stock 3 is the one with the highest variance, the final output should then be:

Date  Stock 3   Stock 2  Stock 1 ...  Stock m
1      300       200      100    ...   500
2      200       300      500    ...   100
:       :         :        :     ...    :
n      400       300      200    ...   100

In order to find the column with highest variance I tried computing:

print(max(df.var()))

However, this only yields the amount of variance without printing the stock ticker. How can I fix this?

CodePudding user response：

You can use .insert() for inserting column at the specified position after you get the locations from .idxmax() and Index.get_loc():

col = df.set_index('Date').var().idxmax()   # get column name of max var

from_pos = df.columns.get_loc(col)          # get location of column with max var

stock1_col = df.columns[1]                  # get column name of first stock
stock1 = df.pop(stock1_col)                 # take out column of first stock

df.insert(1, col, df.pop(col))         # insert column of max var to the original position of first stock
df.insert(from_pos, stock1_col, stock1)      # insert first stock to the original position of column with max var

Result:

(Stock m is the column with max variance in the sample data)

print(col)

'Stock m'



print(df)

  Date  Stock m  Stock 2  Stock 3  Stock 1
0    1      500      200      300      100
1    2      100      300      200      500
2    n      100      300      400      200