SPQR 1.0 (released March 29, 2009) is written in Matlab and can be downloaded from SPQR. The program is for semiparametric quasi-likelihood regression (SPQR) which implements quasi-likelihood for a single index model with unknown link and variance function, for the case of a scalar response and a functional predictor. The user has the option to specify either link function or variance function or both if known. For the case of unknown link function, the model does not contain an intercept parameter and the norm of the parameter vector is constrained at 1. Estimates of link and variance function and of the parameters are provided, as well as an estimate of the covariance matrix. The algorithm uses the iterated reweighted least squares method to minimize the Quasi-Likelihood function and applies kernel smoothing method if the link function and the variance function are unknown. References:

Müller, H.G., Stadtmüller, U. (2005). Generalized functional linear models. Annals of Statistics 33, 774-805.

Chiou, J., Müller, H.G. (2004). Quasi-likelihood regression with multiple indices and smooth link and variance functions. Scandinavian J. Statistics 31, 367-386.

Chiou, J., Müller, H.G. (1998), Quasi-likelihood regression with unknown link and variance functions. J. American Statistical Association 93, 1376-1387.