+61480015851
Econometric Techniques - Analysis on Australian's Marijuana Consumption - Economics Assessment Answer
Download Solution
Order New Solution
Internal Code: 1AGGHB
Economics Assessment Answer
TASK: As a consultant to the Government, you are asked to conduct an analysis on Australians’ marijuana consumption in order to inform relevant policies and practice. You have been provided a sample of individuals from seven waves of a cross sectional drug survey. The surveys were conducted in 2001, 2004, 2007, 2010 and 2016 respectively and the data has been pooled together. The file marijuana#.dta contains data on marijuana participation, (marpart=1 if individual uses marijuana, 0 otherwise), frequency of using marijuana (marfreq = 0 if not current user, marfreq = 1 if used marijuana every few months, marfreq = 2 if used marijuana weekly, and marfreq = 3 if used marijuana daily), year of the survey, respondents’ age (in years and as age categories), gender, marital status, highest level of educational attainment, unemployment status, household income (in logarithm), whether they live in a capital city and whether they reside in a state or territory where possession of small amounts of marijuana has been decriminalised. Precisely, the decrim variable has been coded as a 1 if the individual lives in a State or Territory that has been decriminalised, 0 otherwise. The other variable names are self-explanatory. The dataset also contain information on the price (in logarithm) of marijuana, tobacco and alcohol. Marijuana prices are obtained from the Illicit Drug Reporting System (NDARC 2018), while alcohol and tobacco prices are obtained from the Australian Bureau of Statistics (ABS 2018). A few references to the relevant literature are listed below and you are encouraged to independently source out further literature. You are allowed to undertake the econometric work in small groups, although each one of you will have a different sample. However, it is important that your write-ups are significantly differentiated otherwise you run the risk of plagiarism. The list at the end of the assignment shows the data allocated to each of you. Section A (a) You would like to start by modelling individuals’ decision to use marijuana or not. Based on previous literature, you find that the decision to use marijuana is influenced by demographic and socio-economic factors such as age, gender, marital status, whether they reside in a capital city, unemployment status, their highest level of educational attainment and household income. The price of marijuana and related drugs are also important determinants of the decision to participate. Before estimating the econometric model, conduct a preliminary analysis using descriptive statistics and comment briefly on the results. (Note: since income and drug prices are in logarithm form, descriptive analysis may not be informative). (b) You are not exactly sure what to expect in terms of the direction of the relationship of marijuana participation with age, unemployment status and individual’s level of education. Before estimating the econometric model, conduct a preliminary analysis using cross-tabulations to examine these relationships. Discuss briefly your findings. Give a rationale for entering these variables in the model and why the relationship between marijuana participation and age could be non-linear. Section B (a) Specify a latent model for an individual’s decision to use marijuana. Estimate this using a Probit Model. Present coefficients, average marginal effects and average predicted probabilities. Use the continuous age variable in the model and appropriate time dummies given the data has been pooled across years. (b) Discuss how well the model fits the data overall using a hit-and-miss table. (c) Plot a graph of fitted participation probabilities against age. (d) Another way of examining the non-linear effect of the age variable is to estimate the model with age groups instead of the continuous age variable. Examine the effect of age again using the agegroup variable where the categorical variable is coded as: 1 for 14-17 years, 2 for 18-25 years, 3 for 26-35 years, 4 for 36-45 years, 5 for 46-55 years, 6 for 56-65 years and 7 for above 65 years. Comment on the marginal effects of the age groups. How do these results compare broadly with your findings in c)? (e) Discuss your above results (estimated in part a) in terms of: i) the nature of the economic relationship of marijuana with alcohol and tobacco use (refer to the note on economic relationships below); ii) the demographic and socioeconomic determinants of marijuana use. How do you think the government could use your results in its campaign against marijuana use? (f) Australian States and Territories have gradually eased their laws on marijuana use over the years. Use the decrim variable to examine the impact of decriminalisation on marijuana use and inform the Government. Estimate the model without the year dummies because of the potential collinearity with the decrim variable. Section C Use model specification in Section B, part (a) above to answer the following questions. (a) You can also estimate the decision to use marijuana using a Logit Model. If you do so, which is your preferred model, the Probit or the Logit, and why? [Hint: model selection] (b) It is expected that people who are highly educated behave differently from those with low levels of education. In particular, it is hypothesised that there is greater variability in the decision to use drugs among those who are more educated. Conduct an appropriate test for heteroskedasticity. (c) Estimate an Ordered Probit Model for the frequency of marijuana consumption and the corresponding marginal effects. You are only required to report the marginal effects (not the coefficients) in the report. Discuss the marginal effects of male and education. (d) It is very likely that income is endogenous in this model. In the context of this analysis, discuss what could possibility drive this endogeneity. Section D (a) It is quite likely that the decision to use marijuana and how much (or how frequently) to use are two separate decisions. Discuss an alternative approach for modelling these two decisions giving a rationale for choosing this method. (Note: you are not required to do any estimation). (b) Marijuana is available in four various forms: leaf, head, resin and oil. In the survey individuals also provide information on which type of marijuana they consume. Discuss a modelling strategy for individuals’ choice of marijuana type. (Note: you are not required to do any estimation). (c) Discuss any caveats/issues that may have limited or affected your analysis. You may also like to propose some ideas for further analysis on this issue, such as recommendations for other explanatory factors which could potentially be included in the analysis. Section E Presentation of the Report- Create a heading for each section that suitably reflects its content. ii. To formalise your report, you should write an Executive Summary (100-150 words) at the start of your report which should succinctly outline the purpose of your report and your key findings. iii. Marks will be award for professional presentation of tables and graphs. iv. Include any references in a Reference List/ Bibliography at the end of your report.
This Economics Assignment has been solved by our Economics experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our Experts are well trained to follow all marking rubrics & referencing style. Be it a used or new solution, the quality of the work submitted by our assignment experts remains unhampered. You may continue to expect the same or even better quality with the used and new assignment solution files respectively. There’s one thing to be noticed that you could choose one between the two and acquire an HD either way. You could choose a new assignment solution file to get yourself an exclusive, plagiarism (with free Turnitin file), expert quality assignment or order an old solution file that was considered worthy of the highest distinction