Add Triangle.bfill()

[genedan]

Description

Pandas DataFrame.bfill() can be mirrored to fill in NaNs in a backward direction along the specified axis.

Leaving this as a proposed "Great First Contribution" as this would be easier than the proposed interpolation/extrapolation feature.

Is your feature request aligned with the scope of the package?

• Yes, absolutely!
• No, but it's still worth discussing.
• N/A (this request is not a codebase enhancement).

Describe the solution you'd like, or your current workaround.

tri = cl.Triangle(
    data={
        'origin': [1985, 1985, 1985, 1985, 1986, 1986, 1986, 1987, 1987, 1988],
        'development': [1985, 1986, 1987, 1988, 1986, 1987, 1988, 1987, 1988, 1988],
        'paid': [np.nan, 600, np.nan, 800, np.nan, 1000, 1100, 1200, 1300, 1400]
    },
    origin='origin',
    development='development',
    columns=['paid'],
    cumulative=True
)
tri

          12      24      36     48
1985     NaN   600.0     NaN  800.0
1986     NaN  1000.0  1100.0    NaN
1987  1200.0  1300.0     NaN    NaN
1988  1400.0     NaN     NaN    NaN

tri.bfill(axis=3)

          12      24      36     48
1985   600.0   600.0   800.0  800.0
1986  1000.0  1000.0  1100.0    NaN
1987  1200.0  1300.0     NaN    NaN
1988  1400.0     NaN     NaN    NaN

tri.bfill(axis=4)

          12      24      36     48
1985  1200.0   600.0  1100.0  800.0
1986  1200.0  1000.0  1100.0    NaN
1987  1200.0  1300.0     NaN    NaN
1988  1400.0     NaN     NaN    NaN

Do you have any additional supporting notes?

No response

Would you be willing to contribute this ticket?

• Yes, absolutely!
• Yes, but I would like some help.
• No.

[henrydingliu]

these will all need to be rewritten after we correctly implement zeros. if you have real use cases for these fill functions, then let's prioritize the major refactor around zero v nan. otherwise i don't agree that the scope of this package is to blanket-replicate pandas functionality.