Temple University

Department of Economics

Economics 615

Econometrics I

**Autocorrelation**

The file ICECR.DAT contains 30 observations on variables potentially relevant for modeling the demand for ice cream. This assignment is based on HIildreth and Lu (1960), "Demand Relations with Autocorrelated Disturbances," Michigan State University Agricultural Experiment Station Technical Bulletin 276, East Lansing, MI. Each observation represents a four week period during the years 1951 - 1953. The variables are

Q | Per capita consumption of ice cream in pints |

P | Price per pint in dollars |

I | Weekly family income |

F | Mean temperature in fahrenheit |

(a) Use least squares to estimate the model

(b) Are there any coefficient estimates that are not different from zero? Do you think that the corresponding variables are not relevant for explaining ice cream demand?

(c) Is there any evidence of autocorrelated errors? Check for runs of positive and negative least squares residuals, as well as using the Durbin-Watson and Lagrange Multiplier test statistics.

(d) Estimate the model parameters after transforming the variables to correct for autocorrelation.

(e) Do your estimates from part (d) differ from the results in part (b)?