numpy masked array puzzle

[Tom P]

I have two numpy arrays, xx and yy -
(Pdb) xx
array([0.78200000524520874, masked, masked, 0.37000000476837158,
        0.72500002384185791, 0.60000002384185791, 0.69099998474121094,
        0.78799998760223389, 0.65100002288818359, 0.11100000143051147,
        masked, 0.7160000205039978, 0.54600000381469727,
        0.43599998950958252, 0.63868808746337891, 0.027000000700354576,
        masked, masked], dtype=object)
(Pdb) yy
array([-0.015120843222826226, -0.0080196081193390761, 0.02241851002495138,
        -0.021720756657755306, -0.0095334465407607427,
        -0.0063953867288363917, -0.013363615476044387,
        0.0080645889792231359, -0.0056745213729654086,
        -0.0071783823973457523, -0.0019400978318164389,
        -0.0038670581150256019, 0.0048961156278229494, -0.01315129469368378,
        -0.007727079344820257, -0.0042560259937610449,
        0.0063857167196111056, 0.0024528141737232877], dtype=object)
(Pdb)
-- which gives a strange error -
  stats.mstats.linregress(x, y)
*** AttributeError: 'int' object has no attribute 'view'
(Pdb)

What is stranger I can't get the mask -
(Pdb) np.ma.getmaskarray(xx)
array([False, False, False, False, False, False, False, False, False,
        False, False, False, False, False, False, False, False, False], 
dtype=bool)
(Pdb)