Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

Share this Page URL

Chapter 10. Multiple Regression > 10.1 The Multiple Regression Model - Pg. 298

298 Chapter 10 where Y is a measure of plant growth, X is the amount of nitrogen, and the error term allows for experimental imperfections. Outside the natural sciences, we usually cannot do laboratory experiments under controlled conditions. Instead we must make do with observational data--"nature's experiments"-- and use statistical methods to unravel the effects of one variable on another when many variables are changing simultaneously. In this chapter, we learn how to do this statistical magic. We begin with a description of a regression model with several explanatory variables. Then we see how the model's parameters can be estimated and used for statistical tests. 10.1 The Multiple Regression Model In earlier chapters we looked at Keynes' simple consumption function: C = + Y + where C is spending and Y is household income. It seems plausible that spending is also affected by wealth--households spend more when the stock market or the real estate market is booming and cut back on their spending when these markets crash.