In the text we mentioned that data mining is often used to analyze stock prices and stock returns. A Wall Street Journal article titled “Data Mining isn’t a Good Bet for Stock-Market Predictions” (dated August 8, 2009) by Jason Zweig mentions that data mining has revealed that you can predict stock returns by tracking the number of nine-year-olds in the U.S. Another data mining exercise predicts that stocks are more likely to go up on days when smog is not as bad. While the statistical correlation may be valid, there must be a logical reason why a particular factor will predict stock returns.

Required:

a. Give your opinion on whether the number of nine-year olds in the U.S. should or should not predict stock returns.

b. Give your opinion on whether a smoggy day should or should not predict stock market returns.

c. Give your opinion on what does predict stock returns, if anything.