将每个值乘以两个数据帧，然后将每个答案添加到列中

Question

I am trying to multiply each value in two data frames(df2 and df3) and add each answer to a column in one of the data frames. df2 consists of data where the columns are the types and the rows are how much they cost on average each day. df3 consists of how many of each car is purchased.

I want to multiply each column and row for each day and add it to a column already created in df3 called Money.

I already have a for loop set up and df1 is where I can tell if someone purchased or sold the car and am hoping to somehow incorporate a line in there to loop through each row of df2 and multiply it by df3 and add that amount for each day in the column Money:

for i, rows in df1.iterrows():
    if rows.Buy == 'purchase':
        df3.ix[rows.Day,rows.Type] = row.AmountOfCars
        df3.ix[row.day, 'Money '] -= df3.ix[row.Day,rows.type] * df2???

df1(amount)         Audi    BMW     Buy        
2010-03-16          2000    4000    purchase
2010-04-19          2500    4500    sold

df2(prices)      Audi      BMW   
    2010-03-16  450000    350000    
    2010-04-19  6500000   350300  

df3(totals)      Audi      BMW  Money 
    2010-03-16  2000      4000    ?
    2010-04-19  -2500    -4500    ?

Answer 1

I think you can compare string by str.contains , use loc and sum :

print df1['Buy'].str.contains('purchase')
2010-03-16     True
2010-04-19    False
Name: Buy, dtype: bool

print (df3 * df2).sum(axis=1)
2010-03-16     2300000000
2010-04-19   -17826350000
dtype: int64

df3.loc[ df1['Buy'].str.contains('purchase'), 'Money'] = (df3 * df2).sum(axis=1)
print df3
            Audi   BMW       Money
2010-03-16  2000  4000  2300000000
2010-04-19 -2500 -4500         NaN

If you want only some types of cars use list cars :

cars = ['Audi', 'BMW']
print (df3[cars] * df2[cars]).sum(axis=1)
2010-03-16     2300000000
2010-04-19   -17826350000
dtype: int64

df3.loc[ df1['Buy'].str.contains('purchase'), 'Money'] = (df3[cars] * df2[cars]).sum(axis=1)
print df3
            Audi   BMW       Money
2010-03-16  2000  4000  2300000000
2010-04-19 -2500 -4500         NaN