What is the relationship between crime and punishment? This important question has been examined by Cornwell and Trumbull' using a panel of data from North Carolina. The data are annual for the years 1981 to 1987. The data are in the file Crime Data for Exam.xlsx, which is posted in the Some Datasets folder on the course Contents page. [1] Estimating the Economic Model of Crime with Panel Data," Review of Economics and Statistics, 76, 1994. Consider a regression model in which the crime rate (LN-CRMRTE) is a function of the probability of arrest (LN-PRBARR), the probability of conviction (LN- PRBCONV), the probability of a prison sentence (LN-PRBPRIS), the average prison sentence (LN-AVGSEN), and the average weekly wage in the manufacturing sector (LN-WMFG). The latter represents wages for work that is not illegal. Note that the logarithms of the variables are used in each case. Year is included in the model to see if there is any change in the average crime rate over time. Due to potential nonlinearities indicated in the the scatter plots of some of the independent variables against the dependent, a squared term was added to the model for LN-PRBARR and for LN-PRBCONV. The results are given below: Model parameters (LN-CRMRTE): Standard error Pr>It! Lower bound (90%) Upper bound (90%) Intercept 0.7054 Source Year LN-PRBARR LN-PRBARRA2 LN-PRBCONV LN-PRBCONVA2 LN-PRBPRIS LN-AVGSEN LN-WMFG Value -4.6298 -0.0265 -0.9082 -0.1014 -0.3885 0.0745 0.1618 -0.0848 0.4036 0.0091 0.1260 0.0507 0.0306 0.0177 0.0724 0.0602 0.0684 t -6.5635 -2.9038 -7.2062 2.0588 - 12.6945 4.2144 2.2348 -1.4086 5.9042