Econ 399 A3 Topics covered - University of Alberta



Econ 399 A1/A3 Topics covered by dateDateTopics CoveredReadingsSeptember 1Course outline, course aims, “What is Econometrics?”, examined class example data sets, discussed basic ‘basketball ticket price’ and ‘commodity price’ models.Chapter 1 of TextbookSeptember 3setting up a model for econometrics (dependent variable, explanatory variables, coefficients, random error term);review of functional forms for econometric models (including slopes and elasticities); discussion of preliminary basketball model; introduction to using Stata for graphs, summary statistics and correlations.Chapter 1 of Textbook; functional form handout from web page; First Stata example from webpage.September 8Estimation Methods- “Fitline” example with 10 data points- Derivation of the OLS estimator for a model with only one explanatory variableFitline handout from web page;-Chapter 2.1, 2.2, Appendix 2A in TextbookSeptember 10Using OLS for models with no X’s, one X, multiple X’sUsing Stata’s regress comand for OLS with multiple X’s, using Stata’s predict command to find “y-hat” and “u-hat” Chapter 2.3, 3.1, 3.2 in TextbookFirst Stata example from webpageSeptember 15- Algebraic Properties of OLS- Finding Total, Explained (Model) and Residual Sum of Squares in Stata- Derivation and interpretation of R2 and adjusted-R2- Introduction to statistical properties of OLS: random and non-random components of Y; assumptions about u- Chapter 2.3, 3.1, 3.2, 3.3 in Textbook- Class handout: Review of summation, expectation and variance operatorsFirst Stata example from webpageSeptember 17- statistical properties of u, Y and OLS estimators- (E(β1) in a model with only one explanatory variable;- unbiased estimators-- Var(β1) : formula developed for simple case of only one explanatory variable- Gauss Markov Theorem- omitted variable bias- Chapter 2.5, 3.3, 3.4, 3.5September 22- adding the assumption of normality for the error term- t-tests- Chapter 4.1, 4.2- illustrations from Stata examples 1 and 2September 24- review of t-test procedure- using the lincom command in Stata for t-tests- interpreting p-values- confidence intervals: formula and how to calculate in Stata- Chapter 4.2, 4.3, 4.4- Stata example 2September 29- introduction to F tests: restricted and unrestricted models- F test of overall significance- proof that R-square is zero when there are no X’s in the model - Chapter 4.5- Stata example 2October 1- F tests continued- the relationship between t and F tests- F tests for whether or not a subset of X’s belong in the model- Chapter 4.5- Stata example 2October 6- An introduction to dummy (binary) variables- Allowing for different intercepts across 2 groups- Chapter 7.1, 7,2- Stata example 3- Excel version of example 3 dataOctober 8- F tests for seasonal effects- Chow test (with and without the use of dummy variables)- Chapter 7.3, 7,4, Chapter 10.5 (see sub-section on seasonality)- Stata example 3- Stata example 6October 13- F tests for seasonal effects (same results regardless of “base” period or drop intercept and put in all 4 quarterly dummies)- Reset test for functional form- Using a dummy variable for your “Y”: OLS- Stata example 6- Chapter 9.1- Chapter 7.5October 15- Detailed answers to Part B of Assignment #1 including review of determining whether or not an estimator is unbiased, deriving the OLS estimator for various models, properties of Cobb-Douglas production functions; overview of Part A answers- Part A answers are posted on website; - If you missed class, you will need to get the Part B answers from a classmateOctober 20- Discussed sample mid-term- Discussed answers to Assignment #2- Assignment # 1 returned- sample midterm on website- Assignment 2 answer key on websiteOctober 22MIDTERMOctober 27- Writing Across the Curriculum Presentation (Daniel Harvey)- Review of using OLS with a dummy variable Y (see Oct. 13)- link to presentation slides has been posted on course web pageOctober 29- Measuring prediction success with a dummy variable Y- Probit and Logit as alternatives to OLS- Stata example 4- Chapter 17.1November 3- Midterm returned. (Answers will not be posted. If you missed class you will should get the answers from a classmate.)November 5- Heteroskedasticity (definition, consequence, intro to testing strategies)-Weighted Least Squares- Chapter 8.1, 8.3, 8.4November 17- Heteroskedasticity: how do we detect it (what tests can we use)? (White and BP tests)- Heteroskedasticity: correcting the standard errors, t-tests and F test (using the vce(robust) option in Stata)- Chapter 8.2, 8.3- Example 7November 19- Autocorrelation: Introduction: what is it? what are the consequences?- Rewriting the model (quasi-differencing) to obtain a new equation with no autocorrelationChapter 12.1, 12.3November 24- Autocorrelation tests (BG and Durbin-Watson)- Remedies for autocorrelation (using Stata’s PRAIS and NEWEY commands)- Chapter 12.2, 12.3- Example 8November 26- Empirical Report Workshop - Sample Template posted on class websiteDecember 1- Dynamic ModelsChapter 9.2 (crime rate lagged dependent variable example), 10.2;- Example 9December 3- Multicollinearity and VIFsChapter 3.4, 4.2- Example 9 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download