Selection of a global bandwidth is common in kernel regression. On the other hand, the point wise choice of a local bandwidth can lead to better results in kernel regression because it has direct effect on smoothing of the signal. These smoothing bandwidths affect the filtering capacity of any signals and systems. It shows the higher adaptability to number of analysis varying from various problems in one-dimensional to multidimensional as well as different class of engineering branches of human-machine interactions. In this paper, we have proposed a new method called optimum adaptive local bandwidth selection method (OALB) depending on bias-variance optimization ratio. It is based on Stankovic optimization of bias-variance of the signal [3]. The bandwidth is calculated independently for every points based on intersection of confidence interval.