I have been tasked with taking a set of GPS derived points and doing an analysis on a county wide subset of USPLS Section Lines (technically a polygon layer). I was told to look into Ordinary Least Squares to see which GPS points are more strongly related to the proper positioning of section lines to then serve as a guide for shifting the section lines to a more accurate position. I feel like I have issues with source data from the start as the GPS data was merged into 1 feature class from who knows how many various data sources so the data is not consistently attributed. Also I thought OLS was more along the lines of taking one dataset of various types of data such as (for example) georeferenced crimes and then analyzing things like "offense type" or "time of day" as the dependent variable and then seeing how the other variables impact the dependent. As of now I'm at a loss as to how OLS can help accomplish the section line task put before me. Am I thinking correctly or am I missing something? Any help is very appreciated.