ABSTRACT
A distributed lag model regression occurs when one or more independent variables are lagged one or more periods. There are several justifications for using lagged explanatory variables. The Permanent Income Hypothesis (PIH) offers an interesting application of a distributed lag model. The PIH asserts that spending depends not on actual income but expected future income. The chapter explains autoregressive model regression that uses the lagged dependent variable as an independent variable. It also explains spurious correlation is a strong relationship between variables that is the result of a statistical fluke, not an underlying causal relationship. This occurs often with economic data because economic variables have a tendency to grow over time. Spurious correlations are common with time-series data since so many variables exhibit long-term growth trends. Cointegrated variables can be used as is in regressions without concern for spurious correlation. The chapter describes stationary variable a time-series with constant mean, variance, and correlation between observations any given distance apart.
