Detailed info has been discussed in private messages but a brief outlook again:
Econometric evaluation of an economic issue using data from sources.
-Data set= 3 variables (one dependent, and 2 RHS variables to be added to a constant term).
- multiple regression analysis
-Should include a thorough statistical evaluation using relevaNt diagnositic tests:
Tests such as: T-test, P Test, R squared, Adjusted R squared, Heteroskedacity, dummy variables etc
Would need the following also:
- data used
- do file
- stata output log contaning regression output
Idea is to choose a economic issue, outline a thoery and a set of testable hypothesis that follow from that. Then to test the theory empirically.
Using economic theory to outline a model and its resulting hypothesis to be tested. Set out prior expectatins of the likely sign and magnitudes of coefficients. Also to mention any possible econometric problems likely to be encountered.
To discuss sources of data, giving exact definitions of variables and sample period.
To describe main features of data using table of sample means and standard errors.
To graph the trends and comment on the main trends and features.
To outline the techniques used (ie OLS with correction for autocorrealation or heteroskedacity)
Need to justify why certain techniques are being used
Evaluate model using diagnostic tests (ie box-cox, ramsey rest, forecasting)
Should in tabular form
Based on results