CODEWITHSUNDEEP
python
Jean-Paul Azzopardi
Jean-Paul Azzopardi

Reputation: 381

I just can't figure out this simple calculation in python

I have a df as such:

    Week_Number Year    Open    Close   mean_return labels
0   0   2020    46.270000   46.045000   -0.59950    red
1   1   2020    46.748001   47.506000   1.57500 green
2   2   2020    50.888000   51.382000   1.21760 green
3   3   2020    52.724999   52.530000   -0.20475    red
4   4   2020    49.892001   49.424001   -2.10100    red
... ... ... ... ... ... ...
100 48  2021    123.186000  120.751999  -1.44040    red
101 49  2021    119.786000  119.144000  0.10760 green
102 50  2021    111.006000  108.998000  -1.43240    red
103 51  2021    107.752499  108.429998  0.98400 green
104 52  2021    110.777500  110.529999  -0.05050    red

and I'm trying to calculate how much profit/loss I would make at the end of the 52 weeks according to the labels ('red', 'green').

This is my code so far...


stock = 100 / num

col = 'green'



for n in new_df[2:].itertuples():
    if n[-1] == 'green' and col == 'green':
        col = 'green'
    if n[-1] == 'green' and col == 'red':
        stock = stock * n[3]
        col = 'red'                           
        
    if n[-1] == 'red' and col == 'red':
        col = 'red'
    if n[-1] == 'red' and col == 'green':
        stock = stock / n[4]
        
trading_strategy = stock * new_df.iloc[-2].values[3]

trading_strategy

Essentially what it's saying is, if the week before is 'red' and this week is 'green' then buy, if it is 'green' before and 'red' now sell...

I had run the same script previously and ended up with a value ~ 300 , but now I'm getting 3.132446251028895e-85 and I have absolutely no idea why.

I've been at it for a few hours now and it's 2am, maybe there's something glaringly obvious that I'm not seeing at the moment - any help?

Upvotes: 0

Views: 52

Answers (1)

Macosso
Macosso

Reputation: 1439

You can group by year and compute the sum of returns

df[["Year", "mean_return"]].groupby("Year").sum()

Upvotes: 2

Related Questions