Moving average as a trend generator or smoother for data on a trajectory has been investigated to produce a smooth trend, such that a smooth value can be generated corresponding to every raw data value with the same time or space coordinate. In achieving this goal, computed moving average values are first assigned coordinates at equal intervals such that they cover the same space occupied by raw data. This process is called Stretched Moving Averages. This is followed by a linear interpolation process between stretched moving average values to generate the Stretched Interpolated Moving Averages that have the same coordinates as the raw data values. Accuracy of smoothing obtained depends on the number of data values used in computing the moving averages. This is highlighted by computing the root mean square deviation values, and comparing with the same deviations obtained from local linear regression smoother.