Reference no: EM133684191

Assignment: Introductory Econometrics

Instructions

The coursework is a 1,200 words empirical report.

Your answers should be written in the form of a report of an empirical analysis - as though for publication in a journal. It is an exercise in correctly reporting and interpreting the results. At each stage, you should comment on the model you have estimated. Some examples of things that should be included are:

1) Interpret the coefficients (do they have the signs you expected; are they of a reasonable magnitude, i.e. do they make sense? Remember to use the units of measurement in your answers)

2) Test your coefficients to see if they are significantly different from zero at the 5% and/or 1% level of significance.

3) Assess the goodness of fit of your model, using appropriate statistics, such as the (adjusted) R2or the F-statistic.

Include the output from Stata as an appendix to your report. Include any graphs as an appendix or within the body of the report. Think carefully about the presentation of your work.

Preparations

Each student is supposed to conduct their analysis based on their individual dataset. To obtain your individual dataset, please apply the following steps:

I. Download the dataset Understanding Society.dta, which you can find on Blackboard in the folder "Assignment". It is a pooled sub-sample of the UK Household Longitudinal Study you should be familiar with from exercises in the computing sessions.

II. Open it in Stata. Do not modify it.

III. Run the following lines of code using your 8-digit student ID instead of the placeholder:

set seed student_ID
sample 1

Save the resulting individual dataset. It contains 28 variables and around 800 observations.

IV. Familiarise yourself with your dataset and the included variables. More information can be found in the documentation materials available in the "Assignment" folder in Blackboard.

Task

Using your individual dataset, you will investigate potential determinants of monthly gross wages (paygu_dv) for UK workers. The data comprises details on individuals' labour market activities as well as potentially relevant socio-demographic characteristics. A brief description of the main variables can be found in the table below. More details for all variables can be obtained from the documentation materials or by using the Understanding Society's Variable search function.

Variable name	Short description
paygu_dv	Usual gross pay per month (£)
age_dv	Age (in years)
sex_dv	Sex (binary variable)
hiqual_dv	Highest qualification (categorical variable)
jbft_dv	Full or part-time employee (binary variable)
jbhrs	number of hours normally worked per week
mstat_dv	De-facto marital status (categorical variable)
nchild_dv	Number of own children in household
bornuk_dv	Born in the UK (binary variable)
urban_dv	Living in urban or rural area (binary variable)

a) Explore the data characteristics

Graph paygu_dv against age_dv, sex_dv, hiqual_dv, jbhrsand urban_dv. Choose graphs that are suitable considering the respective variable's scale. Comment on patterns observed and likely model specifications for the relationships. Find any means, standard deviations, correlations etc. that areinteresting and/or useful in the analysis, once again taking the variables' scale into account.

b) Investigate potential wage determinants in a regression framework

Estimate awage regression with paygu_dvas dependent variable, using age_dv, sex_dv, jbhrs, urban_dv and a high education indicator as explanatory variables. Generate the latter based on hiqual_dv, combining the categories "degree" and "other higher" into a high education group, and the other four categories into the reference group.

Review the regression along the lines indicated in the instructions to answer this question.

c) Find the best model

Experiment with different model specifications and find a "best model". A selection of all variables in the data set can be used. To identify a plausible specification, you may wish to resort to economic theory and the existing literature. Note that not all available variables are relevant determinants of wages.

Possibly, a non-linear functional formcan be applied, for example using log-transformations, squared terms, or interactive terms. Model specification tests can be performed and evaluated. Give your arguments why this is the best model. Present only your favourite model in more detail.