Arisaig Final Deliverable

From Analytics Practicum
Revision as of 15:11, 17 April 2015 by Kr.tan.2011 (talk | contribs)
Jump to navigation Jump to search
Appannalogo.png Home Project Proposal Project Management Project Progress Final Deliverable


PROJECT PROGRESS
Final Application


Research has been undertaken to:
1) Understand what factors are of interest in understanding consumption patterns
2) Determine what kind of visualizations are most ideal in displaying the change in trends over time

(1) Factors of Interest in Understanding Consumption Patterns

An article by Minakshi Trivedi [1] discussed analysis performed on changes in consumer behavior patterns of healthy food. The article mentioned that both aggregated and disaggregated data should be taken into account when looking into trends. The article quoted an example that the baby food market (aggregated data) in 2006 saw a fairly flat sales growth from 2005 to 2006 and only a 3.1% increase in 2007. However, when the data is disaggregated, it can be seen that in the same time period, the growth of organic baby food sales saw a 16.4% jump followed by a 21.6% jump in 2007.The increase in popularity did not spill over to potato chips or other snack foods, and was thus unobservable at the aggregate level. Therefore, for our project, we should not only look at overall consumption patterns, but also consumption patterns in smaller product categories.

The same article also found that spatial relationships (spatial correlation, spatial causality or spatial interaction) have a significant importance in the consumption patterns. From their findings, they observed that areas that are in the same cluster produce similar test results. What this means for us is that countries that are geographically close to each other may display similar trends in consumption patterns because “near things are more related than distant things”. There is likely to be more interaction between countries that are geographically close to each other, and hence they are more likely to influence one another. This is an interesting point and we will observe if there is indeed such a pattern shown in our visualizations.