Motivate...
As we near the end of this course, I’d like to provide a bigger picture of how we can utilize all of the tools we’ve learned so far. For this lesson, nothing new will be presented. Instead, we will combine all the statistical metrics we have already learned into one process - a simplified yet effective template for data mining. Please note that there are many ways to go about data mining; this lesson will only present one method.
Before we talk about data mining, I want to present a general motivation for the case study we will focus on. In this lesson, we are going to look at corn. As an agricultural product, you might think that weather (and climate) play a big part in the market. But there are two reasons why this is not necessarily true.
The first is that corn has been genetically altered many times to withstand the extremes of weather. In particular, ongoing research has created drought-tolerant corn hybrids. The second part of this problem is that corn, and in particular corn prices, are largely governed by the futures market. Understanding futures markets is not a requirement for this class, but if we want to ask the right questions, we need to learn a little about all the aspects of the problem. Check out this video that describes the futures markets.
The futures market is regulated more by contracts than by supply and demand. Weather, however, does play an important role (source: "Investing Seasonally In The Corn Market(link is external)"). If we can predict the amount of corn that will be produced or the amount of acreage that will be harvestable (not damaged from weather), that information could be used to help farmers decide how many contracts to sell now or whether a company should wait to buy a contract because the supply will be larger (greater harvest) than anticipated.
Our goal for this lesson is to determine how much weather impacts the corn supply (in the form of harvest, yield, price, etc.), and if certain weather events can be used to predict the corn harvest, providing actionable information on whether to buy or sell a futures contract.
Newsstand
- Global Crop Losses August 2016, Insurers Reel from Losses(link is external) (6:02)
- Corn Field Day Demonstrates Power Of Weather Over Crops(link is external) (3:49)
- Wessell, Maxwell. (2016, November 3). You Don’t Need Big Data — You Need the Right Data(link is external). Retrieved November 18, 2016.
- Brown, Meta S. (2016, October 31). 5 Big Data And Analytics Learning Resources That Most People Miss (But Shouldn't)(link is external). Retrieved November 18, 2016.
- (2016. October 30). How One University Used Big Data To Boost Graduation Rates(link is external). Retrieved November 18, 2016.
Lesson Objectives
- Identify instances when data mining may be useful.
- Describe the general process of data mining.
- Explain what can and cannot be achieved through data mining.
- Interpret results and use for prediction.
