Asymptotic normality of multi-dimension quasi maximum likelihood estimate in generalized linear models with adaptive design

We study the quasi likelihood equation in Generalized Linear Models (GLM) with adaptive design $$\sum\limits_{i = 1}^n {x_i } (y_i - h(x'_i \beta )) = 0$$ ,where yi, is aq-vector, andx i , is ap×q random matrix. Under some assumptions, it is shown that the Quasi-Likelihood equation for the GLM has a solution which is asymptotic normal.