Stata

Homework 2 Eco 480 - Undergraduate Econometrics Question 1 (a) Create the variable race equals to 1 if the individual is white, and 0 if not (black or hispanic). (b) Run the following regression: narr86i = β0 + β1pcnvi + β2avgseni + β3tottimei + β4ptime86i + β5qemp86i + ϵi The results of the regression model is as seen above. The coefficients in the model are as follows; the intercept (_cons) is 0.706. The intercept implies that if all predictors in the model are held constant at 0, the predicted number of arrests in 1986 (narr86) is 0.706. The pcnv coefficient in the model is observed as -0.151 which means there is a decrease in the number of arrests in 1986 by 0.151. This number is statistically significant considering the p-value of pcnv is 0.000 (< 0.05). As for the avgsen coefficient, one month increase in avgsen is linked with a slight decrease in the number of arrest. This observation is similar to ptime86 and qemp86 coefficients whose increase in a month causes decrease in number of arrests when all variables are held constant. Avgsen is linked with a decrease of 0.007, ptime86 cause decrease by 0.039 and qemp86 by 0.103. for avgsen the results are not statistically significant since its p-value is >0.05. The tottime coefficient increases number of arrests by 0.012 when all variables are constant. However, the results of tottime are not statistically significant since it has a p-value that is > 0.05. (c) Calculate the standard error for the estimate of avgsen. As observed above, I used matrix list e(V) to find the variance of avgsen and used Stata to display the square root of the avgsen variance. The result is 0.0124 which is the standard error. (d) Using your results from part (c) compute a 95% confidence interval for the estimate of avgsen. Lower bound is -0.0314 Upper bound is 0.0173 (e) Plot the residuals of the regression. As fitted vales increase, variations are observed in the spread of the residuals. For lower fitted values, there is smaller spread while larger spread is observed as fitted values increase. This suggest heteroskedasticity since the observed residuals do not have constant variance. (f) Create a new variable avgsen2 equals to the average sentence length squared. As observed in the output abo...