Вы находитесь на странице: 1из 24

Lecture 1

The Nature of Econometrics and the Modelling Process


1.1

Aims and Learning Objectives


By the end of this session students should be able to: Appreciate that the study of econometrics is important Appreciate that there are different types of data and different types of variables Understand the nature of the modelling process

1.2

What is Econometrics? The Students Perspective Econometrics is too mathematical; its the reason my best friend isnt majoring in economics The Lecturers Perspective Econometrics allows the measurement and analysis of economic phenomena and the prediction of future economic trends

1.3

What is Econometrics? Historical Context Econometricians wear many different hats Are often criticised for ..using sledgehammers to crack open peanuts while turning a blind eye to data deficiencies and the many questionable assumptions required for the successful application of these techniques (Kennedy,1998: p.2) Econometrics as Art or Science?
1.4

Economic Decisions
To use information effectively: economic theory economic data

economic decisions

*Econometrics* helps us combine economic theory and economic data .


1.5

Types of Data Cross Sectional


Cross-sectional data is a random sample
Each observation is an individual, firm, etc. with information at a point in time If the data is not a random sample, we have a sample-selection problem

1.6

Types of Data Panel

Can pool random cross sections and treat


similar to a normal cross section. Will just need to account for time differences. Can follow the same random individual observations over time known as panel data or longitudinal data

1.7

Types of Data Time Series Time series data has a separate observation for
each time period e.g. stock prices Since not a random sample, different problems to consider These problems will be addressed in EC20019

1.8

The Modelling Process


8 STAGE PROCESS (1) Statement of theory/hypothesis (2) Specification of mathematical model (3) Specification of the econometric model (4) Obtaining the data / conduct preliminary data analysis (5) Estimation of the econometric model and interpretation of regression results (6) Diagnostic Analysis (7) Hypothesis testing (8) Prediction/forecasting
1.9

EXAMPLE Degree Performance in Economics

Number of factors determine performance: Ability Family background Effort Let us look at this from the perspective of ability only and analyse this using a simple bivariate (2 variable) model
1.10

Student Performance

STAGE 1- Statement of Theory /Hypothesis Student Performance Function: Student degree performance is determined by ability

1.11

Student Performance STAGE 2 - Mathematical Model Performance, P, is some function of ability, A :

P = f(A)
In linear form:

(1)

Y = 1 + 2X
where Y = performance and X = ability

(2)
1.12

Student Performance STAGE 3 - Econometric Model

Y = 1 + 2X + U

(3)

Y = Performance - the dependent variable X = Ability - explanatory variable U = Disturbance (random error) term
For this particular example we have data on year 2 average and final year average
1.13

Student Performance STAGE 4a - Obtaining the data Observed values of Y (yr 3 average) and X (yr 2 average) STAGE 4b Preliminary Data Analysis Descriptive Statistics, graphical charts (initial identification of possible errors: outliers, influential observations and lurking variables)
1.14

Student Performance
Year 3 Average Against Year 2 Average
80

70

60

50
Y e ar 3 Av era ge

40

30

20

10

0 10 20 30 40 50 Year 2 Average 60 70 80 90

What can we say about the relationship between year 3 average and year 2 average? Subjective judgement

1.15

Student Performance STAGE 5a - Estimation of the Parameters Y and X are the variables - known 1, 2 and U are the parameters unknown Estimators versus Estimates
The least squares (OLS) regression line is the line that minimises the sum of square deviations of the data points
1.16

Student Performance
Year 3 Average Against Year 2 Average
80

70

60

50 Y e ar 3 Av era ge

40

30

20

10

0 10 20 30 40 50 Year 2 Average 60 70 80 90

1.17

Student Performance STAGE 5b Interpreting the regression results


Check the coefficient sign against expectations (hypothesis) (e.g. do we expect year 2 average to be positively related to year 3 average?) Check coefficient magnitude against expectations (if any) (e.g. are there any prior expectations on the possible magnitude of the effect of year 2 average?)
1.18

Student Performance
STAGE 6 - Diagnostic Analysis Is the model correctly specified?
- Correct functional form - Omitted variables (or unnecessary ones)

Has the model got good diagnostic properties?


(validity of the probability distribution of the disturbance term) - Is the disturbance term uncorrelated with the regressors? - Are the values of the disturbance term independently and normally distributed with mean zero and constant variance
1.19

Student Performance
STAGE 7 - Hypothesis Testing Are the estimates statistically significant? Do they conform with economic theory? STAGE 8 - Forecasting/Prediction For example, predicting the level of performance for a particular ability level. Possible policy implications?
1.20

Student Performance
Excel Regression Output: Bivariate Case
SUMMARY OUTPUT Regression Statistics Multiple R 0.599443 R Square 0.359332 Adjusted R Square 0.356338 Standard Error 4.617821 Observations 216 ANOVA df Regression Residual Total SS MS F Significance F 1 2559.477 2559.477 120.0264807 1.83942E-22 214 4563.393 21.32427 215 7122.87 P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% 4.44707E-21 23.88475288 34.92106 23.88475 34.92106 1.83942E-22 0.441276987 0.6349 0.441277 0.6349

Correlation coefficient

36% of the total variation in year 3 is explained by variation in year 2

Coefficient of determination

Intercept Y2AV

Coefficientstandard Erro t Stat 29.4029 2.799513 10.50286 0.538088 0.049115 10.95566

1 t statistic

P-value

1.21

Student Performance
EViews Regression Output: Bivariate Case
Dependent Variable: Y3AV Method: Least Squares Date: 09/24/05 Time: 20:28 Sample: 1 216 Included observations: 216

1 2

Variable C Y2AV R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat

Coefficient 29.40290 0.538088 0.359332 0.356338 4.617821 4563.393 -635.9494 1.693922

Std. Error 2.799513 0.049115

t-Statistic 10.50286 10.95566

Prob. 0.0000 0.0000 59.87963 5.755834 5.906939 5.938192 120.0265 0.000000

P-value

Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion F-statistic Prob(F-statistic)

Greater number of regression statistics

1.22

The Practice of Econometrics


Economic theory

Econometric model Data Estimation Specification testing and diagnostic testing Is the model adequate? No

Yes Hypothesis testing

Policy: prediction and forecasting

1.23

Summary

In this session we have: Outlined what econometrics is all about Outlined the methodology of econometrics and illustrated the process using a simple economic model

1.24

Вам также может понравиться