Scribd
Upload
71 views2 pages

Ecofmet

Econometrics uses statistics, mathematics, and economic theories to measure and analyze economic phenomena. The methodology involves stating a theory, developing a mathematical model, specifying the model and data, estimating parameters, testing hypotheses, forecasting or predicting, and using the model. The population regression function shows the expected value of Y given values of X variables, while the sample regression function estimates this using least squares. Assumptions of the classical linear regression model include linearity, fixed X values, zero conditional mean error, homoscedasticity, no autocorrelation, zero covariance between errors and X, more observations than parameters, variable X, correct specification, and no multicollinearity. Estimation can occur through least squares, maximum

Uploaded by

Mija Diro
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
71 views2 pages

Ecofmet

Econometrics uses statistics, mathematics, and economic theories to measure and analyze economic phenomena. The methodology involves stating a theory, developing a mathematical model, specifying the model and data, estimating parameters, testing hypotheses, forecasting or predicting, and using the model. The population regression function shows the expected value of Y given values of X variables, while the sample regression function estimates this using least squares. Assumptions of the classical linear regression model include linearity, fixed X values, zero conditional mean error, homoscedasticity, no autocorrelation, zero covariance between errors and X, more observations than parameters, variable X, correct specification, and no multicollinearity. Estimation can occur through least squares, maximum

Uploaded by

Mija Diro
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 2

A. Econometrics literally means economic measurement.

This measurement is frequently undertaken


as a part of an investigation or verification of economic phenomena. Econometrics as a discipline makes
use of an amalgamation of statistics, mathematics and economic theories, concepts or principles.

B.The methodology of econometrics. The use of econometrics as a methodology should involve the
following:

i.)statement of theory or hypothesis

ii.)a presentation of the mathematical [version] of the econometric model

iii.)specification(characterization) of econometric model; inclusive of functional forms,


description of data to be used in the estimation, etc.

iv.)gathering the data to be used in the estimation [inclusive of source and method for
gathering]

v.)estimation of model parameters

vi.)test of hypothesis and other relevant statistical inferences

vii.)forecasting or prediction(if necessary)

viii.)using the model for control(simuation) and policy formulation/implications

C. Regression Functions

The population regression function(PRF) takes the following form

E(Y​i​/X​2i​,…X​ki​) = β​1​ + β​2​X​2i​ +….+ β​k​X​ki

or in stochastic form,

Y​i​ = E(Y​i​/X​2i​,…X​ki​) + u​i

The error term u normally distribution with mean E(u​i​) = 0, variance var(u​i​2​) = σ​2 and covariance
cov(u​i​,u​j​) = 0 for I ≠ j. For the sample regression function(SRF) – using least squares method, takes the
form

Ŷ I​ = β̂ 1​ + β̂ 2​X​2i​ +… + β̂ k​X​ki​ + û i

The residual, û I​ = Y​i​ - Ŷ i​ .

D.Underlying assumptions of classical least square regression model(CLRM)

i.)linearity in parameters

ii.)X​i​’s are fixed


iii.)zero mean; i.e. E(u​i​ / X​2i​,…X​ki​) = 0

iv.)homoscedasticity; i.e. var(u​i​/X​i​) = σ​2

v.)no autocorrelation; i.e. cov(u​i​,u​j​/X​i​) =0 for i ≠ j

vi.)zero covariance; i.e. cov(u​i​ , X​i​) = 0

vii.)number of observations n must be > number of parameters k

viii.)X​i​’s must be variable

ix.)regression model is correctly specified; i.e. correct functional form, no error in


measurement, etc.

x.)no multicollinearity; i.e. cov(X​i​ , X​j​) = 0 for i ≠ j

n
E.)3 possible methods of estimation: least squares(min ∑ û i​2​), maximum likelihood(estimating
i=1
unknown parameters while maximizing probability of estimating Y high) and method of
moments(population characteristic is estimated by using sample characteristic).

F.)There are desirable properties of estimators: unbiasedness E( θ̂ ) – ​𝛉 = 0, efficiency(minimum


variance when compared with alternative estimators), linearity(linear function of sample observations
and consistency(estimator approaches parameter as sample size increases indefinitely)

You might also like