In recent years, considerable effort has been spent in developing high adaptive array superresolution techniques. This paper presents a perturbation expansion method named "two-variable technique" to derive an expression for the output power of the adaptive null-steering antenna array considering the statistical weight fluctuations. This array utilizes Howells-Applebaum type adaptive weight. Thus, in turn, the paper presents the difference between the assumed and actual adaptive array signal processing response of different high resolution techniques. These techniques are based on maximum-likelihood method, adaptive-angular response, and thermal noise algorithm. By evaluating the model performance with different signal-to-noise ratios, number of antennas, and different signal arrival angles for selected test cases, we gain insights into the sensitivity of the signal estimation problem to model uncertainty.