The paper “ Applied Econometrics for Economics and Finance” is an informative variant of the assignment on macro & microeconomics. The equation shows that when the number of cigarettes is increased log(income) is increased. Looking at p- values of the coefficients only Cigs are not statistically significant using a confidence level of 0.05 since 0.4754> 0.05 while the remaining variables, which are education, Age square, age are statistically significant since the p-value is 0.00 < 0.05 hence statistically significance. Age square is inversely related to income while the rest are positively related. In the case of a casual relationship between variables are presented by the structural equation, the best percentage approximation will be given by 100.β 1 which is the percentage change in income in the case of a person who smokes one more cigarette per day (Winkelmann 2013). Looking at the graphs shown below, it will be noted that each variable exhibits different patterns throughout the analysis explaining the independent characteristics of each variable. The graphs of each individual variables are shown in the figure below 2.0Question two If white is added in the equation, the results are shown below From the descriptive statistics, the number of cigarettes smoked per day coefficient is 0.0002 with a confidence level of 0.2956.
The other variables have their coefficients as shown in the equation below The equation shows that when the number of cigarettes is increased log (income) is increased. Looking at p- values of the coefficients only Cigs are not statistically significant using a confidence level of 0.05 since 0.2956> 0.05 while the remaining variables, which are education, Age square, age, white are statistically significant since the p-value is 0.00 < 0.05 hence statistically significance. When white people are added, nothing changes except the number of cigarettes is increased but it remains statistically insignificant while the rest remain the same (Glickman 2014). 3.0Question three Residual graphs From the graph, the presence of heteroskedasticity does not exist in the residual graphs due to equal distribution and patterns in the data hence there is no problem of heteroskedasticity. (b) LM test
Chen, H., Cummins, J.D., Viswanathan, K.S. and Weiss, M.A., 2014. Systemic risk and the interconnectedness between banks and insurers: An econometric analysis. Journal of Risk and Insurance, 81(3), pp.623-652.
Glickman, N.J., 2014. Econometric analysis of regional systems: explorations in model building and policy analysis. Academic Press.
Lunde, A., Shephard, N. and Sheppard, K., 2015. Econometric analysis of vast covariance matrices using composite realized kernels and their application to portfolio choice. Journal of Business & Economic Statistics, (just-accepted), pp.1-46.
William, H., 2012. Econometric Analysis Solutions Manual 5th ed.
Winkelmann, R., 2013. Econometric analysis of count data. Springer Science & Business Media.