Practical Econometrics I / II

[PE I] 2019 Fall Semester

2019-08-17 10:27

RESULTS: [Link]


Announcements

  • Extra Task(-s): Extra task file added. Note that each group is assigned a topic and a dataset. The extra task is preliminary worth at least 5% (that is equivalent to 0.5 to the final grade).
  • Midterm I: results added (see above file);
  • Midterm II (with R):
    • Date: 2019 – 11 – 28
    • Time: [12:00 – 15:00+]
    • Location: Computer Classes (Šaltinių g., 2nd floor) [203207]
    • (optionally – from 10:15+ to 12:00 in 301 and the remainder in [203207 in Šaltinių g., 2nd floor]
  •  The lecture on 2019 – 12 – 12 will be moved to an earlier (TBA) date (preliminary – either to 2019 – 12- 09 from 14:20, or, alternatively +2h to lectures on 2019-11-28 and 2019-12-05)

 

  • Added Template files (with some formula/matrix examples) – this is to get a general idea on how you can format your code and your insights based on the code and its output in a way, that is clear and readable.

Important! Download the following [.ipynb file]  file and make sure that everything works on your machine.

ERASMUS students, who want to use R + RStudio, can use RStudio and verify that everything works by downloading the [.Rmd] file and inputting their student code and pressing the “Knit” button in RStudio to generate a .html file.

 


Course Information

Important: from 2019 the course is taught in English.

 TitleLink
1Course Introduction, Overview and RequirementsLink,
R_vs_Python
Setup R, Python and JupyterLab on MIF Linux computersLink
2Lecture Notes (usually updated throughout the course)Link
3Template files on how to separately write down various formulas, as well as format text/comments alongside your code blocks/chunks and its output.

The code is provided only as a simple example to see how it visually looks like alongside some random comments.
Python: JupyterLab
(HTML output)

R: JupyterLab
(HTML output)

R: RStudio
(HTML output)

 


Lecture Files

 TitleLink
1Univariate Regression:
General Concepts, OLS, R & Python implementation
Link
2Univariate Regression:
OLS, Regression Models & Interpretation
Link
3Univariate Regression:
MLE, Confidence Intervals and Hypothesis Testing
Link
4Univariate Regression:
Hypothesis Testing (Review), OLS Prediction and Prediction Intervals, GoF
Link
📊Univariate Regression:
Data Subsampling & Chapter Review
Link
5Multivariable Regression:
Model Specification
Link
6Multivariable Regression:
OLS, Confidence Intervals, Hypothesis Testing & GoF
Link
7Multivariable Regression:
Restricted Least Squares, Multicollinearity
Link
8Multivariable Regression:
Generalized Least Squares, Heteroskedastic and Autocorrelated Errors
Link
9Multivariable Regression:
General Modelling Difficulties
Link
📊Multivariable Regression:
General Modelling Difficulties
10Discrete Response Models:
Binary Response Variables - Logit, Probit & Parameter Interpretation
11Discrete Response Models:
Binary Response Variables - Goodness-Of-Fit
12
13
14
EXTRA TASKLink

 


Some Useful Links

 TitleLanguageLink
1NIST/SEMATECH e-Handbook of Statistical MethodsEnglishLink
2Principles Of Econometrics with REnglishLink
3Using R for Introductory EconometricsEnglishLink
4An Introduction to Statistical Learning with Applications in REnglishLink (Homepage)
5The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2nd Ed.)EnglishLink (Homepage)

Note: in the past, there were free DataCamp courses available (as introductions to R and Python programming languages). Unfortunately, as of around Q1 of 2019 this is no longer the case.

Consequently, the lecture notes are updated to include links to alternative freely available interactive courses for R (using `swirl` package) and Python (using PyCharm Edu) programming languages.


[PE I] 2018 Fall Semester

2018-06-21 16:53

Announcements

  • Rezultatai įkelti į VU sistemą.

 

Rezultatai: [RESULTS], [EXTRA_TASK_RESULTS]

 


Course Information

 TitleLink
1Course IntroductionIntroduction
2Literature (Theory)R.Lapinskas, Practical Econometrics I. Regression Models (Lecture Notes)
3Literature (Practice)R.Lapinskas, Practical Econometrics I. Regression Models (Computer Labs)

[(LINK) A First Lesson In Econometrics (2 pages)]


Lecture Files

 TitleLink
1Lecture Notes (updated for each lecture)Link
2Univariate Regression TasksLink
Univariate Regression Example with wage dataset
Other Univariate Regression Examples
[HTML (R)] [.ipynb] [.7zip]
Link
3Multiple Regression Tasks (updated for each lecture)Link
4Multiple Regression Example with wage dataset[HTML (Python)], [.zip], [.ipynb]
[HTML (R)], [.ipynb]
5Discrete Response Model TasksLink
6Discrete Response Model Example with labor force participation dataset[HTML (R)], [.ipynb]
7EXTRA TASKLink

 


Some Useful Links

 TitleLanguageLink
1Econometrics website, literature, dataset links, etc.LithuanianLink
2A. Račkauskas. Įvadas į ekonometriją (Paskaitų konspektai)LithuanianLink
3Terminology and overview of descriptive statistics and probability theoryLithuanianLink
4NIST/SEMATECH e-Handbook of Statistical MethodsEnglishLink
5Principles Of Econometrics with REnglishLink
6Using R for Introductory EconometricsEnglishLink

 


Additional Info

 

  • Examine the example the [.ipynb] file, to make sure that you can run code from online python script files (a similar system will be used for Midterm II)

To use Cramer-von-Mises test in Python:

  • Open Anaconda Navigator. Go to “Environments”;
  • Click on the green arrow next to “base (root)” and select “Open Terminal”;
  • Enter the following command to install the required package:

pip install scikit-gof

  • To use the test for a fitted model  variable “my_model” write:

import skgof as skgof

skgof.cvm_test(data = my_model.resid, dist = stats.norm(0, np.sqrt(np.var(my_model.resid))))[1]