A Convergence Theorem for Random Linear Combinations of Independent Normal Random Variables
Christopeit, N. ; Helmes, K.
Ann. Statist., Tome 7 (1979) no. 1, p. 795-800 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
It is shown that under fairly mild conditions linear combinations of independent normally distributed random variables with random coefficients tend to zero almost everywhere. The result is applied to parameter estimation in linear regression models.
Publié le : 1979-07-14
Classification:  Strong law of large numbers,  regression analysis,  60F15,  60G50,  62J05 
@article{1176344729,
     author = {Christopeit, N. and Helmes, K.},
     title = {A Convergence Theorem for Random Linear Combinations of Independent Normal Random Variables},
     journal = {Ann. Statist.},
     volume = {7},
     number = {1},
     year = {1979},
     pages = { 795-800},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176344729}
}

Christopeit, N.; Helmes, K. A Convergence Theorem for Random Linear Combinations of Independent Normal Random Variables. Ann. Statist., Tome 7 (1979) no. 1, pp.  795-800. http://gdmltest.u-ga.fr/item/1176344729/