site
stats
Making Count-Based Claims from Sample Data (2019) | Ifpress.com

Discovering Time-Dependent Trends

Lampe, R. and Ennis, D. M. (2007). IFPress, 10(1) 2-3.

Abstract:

 

Product changes in manufacturing and storage are often time-dependent. Sometimes these changes are linear, sometimes cyclical and at other times chaotic. It is often not obvious that a time-dependent change is in fact occurring in some variable of interest. In this report we discuss methods for studying whether a trend exists and, if so, what the nature of that trend is. Data used to study this problem, observations collected sequentially over time, are referred to as a time series. Time series techniques can prove to be important in bringing to light underlying trends in data that may be evident through other techniques.

Figure 2. Fourteen-point moving averages of lemon juice color values plotted against chronological order of observations.

Follow us for updates on the latest IFP news!

  • Grey LinkedIn Icon

Copyright © 2019 The Institute for Perception. All rights reserved. Please review our privacy policy prior to accessing our website.