# Regression analysis for Measurement values

I'm in a situation where I'm provided with a signal and a couple of measurement values for that signal. Usually the signal will drift a little when the measurement start (i.e. because of heat). I want to determine whether my system is still drifting (at any given point) or not. Therefore I wanted to determine a regression line and used the equations for linear regressions (https://en.wikipedia.org/wiki/Simple_linear_regression), but the result is awkward and cannot be correct. So the question is, what am I missing?

• Can you elaborate why it is awkward? – Alberto Takase Feb 7 at 16:14
• I think the problem is my signal is drifting at the beginning and constant from a certain point onwards (a little noisy), but it's not linear dependant on the time, right? – user3292642 Feb 7 at 16:16
• I recommend that you provide more details about what your specific question is. It's hard to say if anything is missing at all based on the information you provided. – Alberto Takase Feb 7 at 16:24