df1:
| name_d | name_o | year |
|---|---|---|
| Turkiye | Italy | 1990 |
| Turkiye | Italy | 1991 |
| Turkiye | Italy | 1993 |
| Spain | Italy | 1990 |
| Spain | Italy | 1991 |
| Spain | Japan | 1990 |
df2:
| country_name | year | v2x_regime |
|---|---|---|
| Spain | 1990 | 0 |
| Turkiye | 1990 | 0 |
| Italy | 1990 | 1 |
| Turkiye | 1991 | 1 |
| Spain | 1991 | 1 |
| Italy | 1991 | 1 |
Expected result:
| name_o | v2x_regime_name_o | name_d | v2x_regime_name_d | year |
|---|---|---|---|---|
| Italy | 1 | Turkiye | 1 | 1990 |
Basically I would like to know the regime type of the each country for each year. Since this is a bilateral data, there are two columns that include country name. For example, for each year I would like to have the index for name_o column and name_d column.
CodePudding user response：
Does this work for you?:
df = df1.merge(df2, left_on=['name_o','year'], right_on=['country_name','year'], how='left')
df = df.merge(df2, left_on=['name_d,'year'], right_on=['country_name','year'], how='left')