In *The resistant line* section, we constructed our first regression model for the relationship between two variables. In some instances, the *x*-values are so systematic that their values are almost redundant, and yet we need to understand the behavior of the *y*-values with respect to them. Consider the case where the x-values are equally spaced; the shares price (*y*) at the end of day (*x*) is an example where the difference between two consecutive *x*-values is exactly one. Here, we are more interested in smoothing the data along the *y*-values, as one expects more variation in their direction. Time series data is a very good example of this type. In the time series data, we typically have *x _{n + 1}*

Start Free Trial

No credit card required