We develop a conditional factor model for the term structure of treasury bonds, which unifies non parametric curve estimation with cross-sectional asset pricing. Our factors correspond to the optimal non-parametric basis functions spanning the discount curve. They are investable portfolios...

This paper presents constructive methods of factor neural network (FNN) and their applications to the study of intelligent computation. Liu Zengliang is the first to apply this theory it to intelligent computation, it allows describing numerically and simulating such intelligence problems as...

It is known that the principal component estimates of the factors and the loadings are rotations of the underlying latent factors and loadings. We study conditions under which the latent factors can be estimated asymptotically without rotation. We derive the limiting distributions for the...