[BUG] WeightRegression produces incorrect regression when thru_orig = False

[henrydingliu]

Are you on the latest chainladder version?

• Yes, this bug occurs on the latest version.

Describe the bug in words

came across a bug in _fit_OLS while trying to recreate chapter 11. likely culprit is a missing w (or two) around here. should be a quick PR for the students in our contributor group

How can the bug be reproduced?

test = cl.Triangle(
    data = {
        'value':[100,200,300,400,500,0,0,0],
        'weight':[1,1,1,1,1,0,0,0],
        'ay': [2001,2002,2003,2004,2005,2006,2007,2008],
        'dev':[2008,2008,2008,2008,2008,2008,2008,2008],
    },
    origin = 'ay',
    development='dev',
    columns=['value','weight'],
    cumulative=True,
)
reg = cl.WeightedRegression(
    axis=2,
    thru_orig=False,
    xp=test.get_array_module()
).fit(
    np.ones((1,1,8,1)) * test.origin.year.values[None,None,:,None],
    test['value'].values,
    test['weight'].values
)
reg.slope_

What is the expected behavior?

100

Would you be willing to contribute this ticket?

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