Fill nan values from a column with values from another column-CodePudding

I'm working with a time-series dataset whose my interest columns are the month column (mes), the year column (ano), the minimum prices (min) and the modal price (mcom).

some months from the year 2013 doesn't have any value in the min column, and in those months and want to fill with the values that are in the mcom column.

I used the following line to find the months where there are no min values:

dfc.loc[(dfc['ano'] == 2013) & (dfc['mes'] == 1)]

Basically, I want to fill the 'min' column with the values in the 'mcom' column when

dfc['ano'] == 2013 & dfc['mes'] == 1

how can I do it?

That's the head of the dataframe:

data    dia mes ano wday    prod    und proc    tipo    min mcom    max merc    date    julian
6   02/01/2013  2   1   2013    quarta  Peixe de agua salgada   Kg  RS-SC   Sardinha    NaN 5.28    5.28    Est 2013-01-02 12:00:00 1.416667
14  03/01/2013  3   1   2013    quinta  Peixe de agua salgada   Kg  RS-SC   Sardinha    NaN 5.28    5.28    Est 2013-01-03 12:00:00 2.416667
22  04/01/2013  4   1   2013    sexta   Peixe de agua salgada   Kg  RS-SC   Sardinha    NaN 5.28    5.28    Est 2013-01-04 12:00:00 3.416667
30  07/01/2013  7   1   2013    segunda Peixe de agua salgada   Kg  RS-SC   Sardinha    NaN 5.28    5.28    Est 2013-01-07 12:00:00 6.416667
38  08/01/2013  8   1   2013    nan Peixe de agua salgada   Kg  RS-SC   Sardinha    NaN 5.28    5.28    Est 2013-01-08 12:00:00 7.416667

The 'tipo' variabl, only has 'sardinha' and the types of the variables are:

dfc.dtypes

data       object
dia         int64
mes         int64
ano         int64
wday       object
prod       object
und        object
proc       object
tipo       object
min       float64
mcom      float64
max       float64
merc       object
date       object
julian    float64
dtype: object

CodePudding user response：

m = (dfc['ano'] == 2013) & (dfc['mes'] == 1)

dfc.loc[m, 'min'] = dfc.loc[m, 'mcom']
# or
dfc['min'] = dfc['min'].mask(m, df['mcom'])
# or
dfc['min'] = np.where(m, df['mcom'], dfc['min'])