Macroeconomic Analysis Maria Vyshnya icon

Macroeconomic Analysis Maria Vyshnya




НазваMacroeconomic Analysis Maria Vyshnya
Дата29.06.2013
Розмір445 b.
ТипДокументи


Macroeconomic Analysis

  • Maria Vyshnya

  • maria.vyshnya@gmail.com

  • (office hours – upon request)


Slide 2 What is different:

  • Topic 1 “Purpose and instruments…” – at the end of the course

  • Topic 6 (?) “Indicators of Macroeconomic Disproportions”: “Corruption” is taught instead of the mentioned topic

  • Additional literature might be recommended over the course of trimester

  • Sequence of topics might be changed. First five topics:

  • Econometrics outline

  • Soft budget constraints (МБО)

  • Shadow Economy

  • Investment climate



Slide 3 Important:

  • Trimester grade points - 60:

  • Quizzes – 32 points (8 quizzes, 4 points per quiz)

  • Interim – 28 points

  • “pass grade” – 30 (minimum)

  • Final exam: 40 points

  • “pass grade” – 20 (minimum)

  • Inform me about emergency cases immediately!!!!



Slide 4 Important:

  • September 12-22 I am out of Ukraine.

  • “Compensating” lecture is to be scheduled later



Slide 5 Topic 1: Econometrics Outline

  • Purpose:

  • Refresh the very basic information crucial for the course understanding

  • Lecture plan:

  • What is econometrics

  • Structure of economic data

  • Functional form selection

  • Test of the model and covariates “goodness of fit”



Lecture plan

  • What is econometrics

  • Structure of economic data

  • Functional form selection

  • Test of the model and covariates “goodness of fit”



Slide 6 What is Econometrics

  • The simplest possible econometric model:

  • y= β0 + β1X1 + … + βnXn + u (1)



Slide 7 What is Econometrics

  • Imagine you are hired by the government to evaluate the effectiveness of a publicly funded job training program. This program addresses unemployed workers registered in local unemployment offices, and teaches them various ways to use computers. The twenty-week program offers courses, and enrollment in all or part of the program is voluntary. You are to determine what, if any, effect the training program has on each worker’s subsequent hourly wage.

  • (modified Wooldridge J. Introductory Econometrics: A Modern Approach. Third edition. Thompson press)



Slide 8 What is Econometrics

  • Simple job training model:

  • y= f (education, experience, training) (2)

  • or

  • y= β0 + β1 education + β2 experience +

  • + β3 training + u (3)

  • u (hidden person-specific factors): talents of a person, quality of education, family background etc.





Lecture plan

  • What is econometrics

  • Structure of economic data

  • Functional form selection

  • Test of the model and covariates “goodness of fit”



Slide 10 Structure of Economic Data

  • Simple linear regression model:

  • y= β0 + β1X + u (4)

  • Multiple linear regression model:

  • y= β0 + β1X1 + … + βnXn + u (5)









Slide 14 Structure of Economic Data

  • Panel Data:

  • yit= β0 + β1X1it + β2X2it +…+ βnXnit + uit, (8)

  • where

  • i – an “individual i”, i=[1,m]

  • t – time “t”, t=[1,p]

  • For example:

  • yit= β0 + β1educit + β1experit + β1trainingit + uit, (9)



Lecture plan

  • What is econometrics

  • Structure of economic data

  • Functional form selection

  • Test of the model and covariates “goodness of fit”





Slide 16 Functional Form Selection

  • “Mixed” (often used in real life):

  • ln y= β0 + β1ln z1 + β2w + β3X + β4X2 + u (10)

  • β1 - per cent change in z

  • β2 - unit change in w

  • Logs are NOT used:

  • Variables are measured in years (e.g. education, experience, tenure, age)

  • Variables that are proportions or per cent (e.g. unemployment rate, percentage of students passing an exam



Slide 17 Functional Form Selection

  • ln y= β0 + β1ln z1 + β2X + u (11)

  • Example 1.1:

  • (Wooldgridge J. Introductory Econometrics: 3d Edition, Example 4.5, p.139) :

  • Log(price) =11.08 - 0.954 log(nox) + 0.255 rooms, (12)

  • where:

  • β1 – when nox increases by 1%, price falls by 0.954% (holding rooms fixed!!!)

  • β2 – when rooms increase by 1 (unit/room), price increases by 100*(0.255)=25.5% (holding nitrogen oxide fixed!!!)



Slide 18 Functional Form Selection

  • In (12):

  • Log(price) =11.08 - 0.954 log(nox) + 0.255 rooms, (12)

  • Alternatively:

  • Price =1304 - 83000 log(nox) + 13500 rooms, (13)

  • where:

  • β2 – when rooms increase by 1 (unit/room), price increases by 13500 hryvnya (holding nitrogen oxide fixed)

  • β1 – when nox increases by 1%, price falls by 83000/100=830 hryvnya (holding rooms fixed!!!)



Slide 19 Functional Form Selection

  • Quadratics:

  • ln y= β0 + β1X + β2X2 + u (14)

  • The critical/turning point is calculated using first derivative.

  • If the critical point exceeds the maximum value of an independent variable (or is less then the minimum point):

  • the “non-trivial” values of an independent variable are to be ignored; however, the effect of X on Y is increasing (diminishing) rather than linear

  • The estimated effect is biased, since the model is missing some important variables



Slide 20 Functional Form Selection

  • Assignment 1.1:

  • Log(price) =13.39 - 0.902 log(nox) – 0.087 log(dist)-

  • - 0.545 rooms + 0.062 rooms2

  • (15)

  • Example 1.2:

  • (Wooldgridge J. Introductory Econometrics: 3d Edition, Example 6.2, p.202)



Slide 20 Functional Form Selection

  • Assignment 1.2:

  • Choose the proper functional form for a job training program model (discuss different options):

  • Wage = f (education, experience, training)



Lecture plan

  • What is econometrics

  • Structure of economic data

  • Functional form selection

  • Test of the model and covariates “goodness of fit”



Slide 21 Test of the Model and Covariates “Goodness of Fit”

  • Significance of an individual coefficient:

  • ln y= β0 + β1ln z1 + β2w + β3X + β4X2 + u (10)

  • Hypothesis 1: β1 = 0

  • Hypothesis 2: otherwise

  • Test Hypotheses using:

  • t-statistics

  • p-value (hint: p-value intuitively reflects the probability of a mistake – an “insignificant” variable is included in a model)

  • A coefficient is significant, if

  • p-value<0.01 (1% of mistake);

  • p-value<0.05 (5% of mistake); or

  • p-value<0.1 (10% of mistake);



Slide 22 Test of the Model and Covariates “Goodness of Fit”

  • Significance of several coefficients:

  • General Walt test (read Wooldridge if needed)

  • Overall significance of a model:

  • R2 - for linear models (R2 >0.8 is a strong evidence of a useless model!!!)

  • Akaike, Schwarz criteria (lower values - better models)

  • Specific Walt test (read Wooldridge if needed)

  • Comparing drastically different models (e.g. a cross section and a panel model; OLS and fixed effects models etc)

    • Drop one or two latest observations
    • Run each of the model
    • Predict the dropped observations
    • Check which model fits the dropped values better


Slide 23 Test of the Model and Covariates “Goodness of Fit”

  • Goodness of fit tests (all of them!) are invalid if there is heteroskedasticity or serial correlation.

  • Ideal modeling:

  • Test for heteroskedasticity (Breusch – Pagan test, White’s test), correct heteroskedasticity

  • Test for serial correlation (not applicable to cross-section models!!!) (Durbin - Watson test – for the first order serial correlation only!; Lagrange Multiplier test – universal), correct serial correlation

  • Run “goodness of fit” tests



Slide 24 Test of the Model and Covariates “Goodness of Fit”

  • Problem Set 1.2:

  • In 2002 Ministry of Health of Ukraine has initiated “Maternal and Infant Health Project” (MHIP) in 20 pilot maternities. The project is supposed to implement modern perinatal technologies aimed to improve maternal and infant health characteristics.

  • Currently, the first phase of the project has finished. If the first phase is successful, the Ministry would be willing to expand the project and to adopt their standards for the entire country.

  • Your assignment:

  • compare MIHP and non-MIHP rayons of UA, and

  • evaluate efficiency of the first stage of the project using different medical and socio-economic characteristics of rayons.



Схожі:

Macroeconomic Analysis Maria Vyshnya iconPrivatization in uk macroeconomic Analysis

Macroeconomic Analysis Maria Vyshnya iconShadow Economy Macroeconomic Analysis

Macroeconomic Analysis Maria Vyshnya iconPrivatization in Ukraine Macroeconomic Analysis

Macroeconomic Analysis Maria Vyshnya iconForeign Direct Investments (fdi) Macroeconomic Analysis
Білоцерковський О. та ін. Інвестування в Україні як процес відбору. // Welcome. 2002, №7. с. 3-6
Macroeconomic Analysis Maria Vyshnya iconSoft Budget Constraints Macroeconomic Analysis
Легейда Н. Непрямі субсидії в Україні… Rodrik D. Industrial Policy for the Twenty-First Century
Macroeconomic Analysis Maria Vyshnya iconKozlova Maria O

Macroeconomic Analysis Maria Vyshnya iconMaria januszewska-warych bologna process. Hopes and concerns

Macroeconomic Analysis Maria Vyshnya iconMobility Circle and Return Migration a case study of Bulgarian Students Maria Velizarova

Macroeconomic Analysis Maria Vyshnya iconAnalysis of pharmacies financial condition analysis of financial stability financial stability of the enterprise

Macroeconomic Analysis Maria Vyshnya iconMacroeconomic problems The role of the social sphere in the reproduction of human capital

Додайте кнопку на своєму сайті:
Документи


База даних захищена авторським правом ©zavantag.com 2000-2013
При копіюванні матеріалу обов'язкове зазначення активного посилання відкритою для індексації.
звернутися до адміністрації
Документи