Scribd
Upload
16 views5 pages

ECON1150 Lec 02

This document discusses key concepts in econometrics including: 1) Linear regression models relate a dependent variable to one or more independent variables plus a stochastic error term. 2) Least squares estimation is used to estimate the coefficients by minimizing the residual sum of squares. 3) The coefficient of determination measures how much of the variation in the dependent variable is explained by the regression model. 4) Hypothesis testing using t-tests and F-tests evaluate the statistical significance of estimated coefficients and the overall model. 5) Dummy variables are used to capture the effect of categorical variables in a regression.

Uploaded by

vietthu35
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
16 views5 pages

ECON1150 Lec 02

This document discusses key concepts in econometrics including: 1) Linear regression models relate a dependent variable to one or more independent variables plus a stochastic error term. 2) Least squares estimation is used to estimate the coefficients by minimizing the residual sum of squares. 3) The coefficient of determination measures how much of the variation in the dependent variable is explained by the regression model. 4) Hypothesis testing using t-tests and F-tests evaluate the statistical significance of estimated coefficients and the overall model. 5) Dummy variables are used to capture the effect of categorical variables in a regression.

Uploaded by

vietthu35
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

AOF – IIFE – ECON1150 – Econometrics for Finance

LECTURE 02
ECONOMETRICS

Variable

Y X (X 1 , X 2 , … ¿

Dependent Independ
Explained Explanatory
Predicted Predictor / Predictand

Page 1 of 5 Le Thanh Ha
AOF – IIFE – ECON1150 – Econometrics for Finance

LINEAR REGRESSION MODEL


Model for Time series
Population regression model: Y t =β 1+ β2 X 2 t + …+ β k X kt +u t
Or on average: E(Y ¿¿ t)=β 1 + β 2 X 2 t + …+ β k X kt ¿

Sample regression model: Y t = ^β 1+ ^β2 X 2 t + …+ ^β k X kt + u^ t

Or on average: Y^ t = ^β 1+ ^β2 X 2 t + …+ ^β k X kt

 β j : coefficients
 β 1: intercept
 β j ( j=2 ÷ k ): slope, partial coefficient of X j
 ut : random errors
 ^β : estimated coefficient, estimator
j

 u^ t : residuals

FORM OF EQUATION

Lin-lin Y t =β 1+ β2 X t + ut X increases 1 unit  Y increases β 2 unit

Log-log ln Y t =β 1+ β2 ln X t +ut X increases 1%  Y increases β 2%

Lin-log Y t =β 1+ β2 ln X t + ut 1
X increases 1%  Y increases β %
100 2
Log-lin ln Y t =β 1+ β2 X t + ut X increases 1 unit  Y increases 100 β 2%

LEAST SQUARE (LS) METHOD


Find ^β j ( j=1 ÷ k ) that RSS=∑ u^ 2t → min❑

Gain ^β j , SE ( ^β j ) , Cov( β^ j , ^β i)

Page 2 of 5 Le Thanh Ha
AOF – IIFE – ECON1150 – Econometrics for Finance

Satisfy LS assumptions then ^β j are BLUE (Best Linear Unbiased Estimator)


COEFFICIENT OF DETERMINATION

TSS Total Sum of Squares Variation of dependent variable


ESS Explained Sum of Squares Variation that explained by model
RSS Residual Sum of Squares Variation that explained by random errors

2 ESS RSS
R= =1− : coefficient of determination
TSS TSS
Meaning: Proportion (%) of total variation of dependent variable that explained by model
Adjusted R2: R2 : Adj R-sq < R-sq

CONFIDENT INTERVAL
Estimate for coefficient:
^β −SE ( ^β )∙ t ^ ^
j j ( n−k ) α /2 < β j < β j + SE( β j )∙t ( n−k ) α / 2

T-TEST FOR COEFFICIENT

Statistic value Hypothesis Critical value Reject H 0


^ β
β− H 0 : β=β 0
t= 0
t (n−k ) α /2 |t |>t (n−k ) α / 2
^
SE ( β) H 1: β ≠ β0

H 0 : μ=μ 0
t (n−k ) α t >t ( n−k ) α
H 1: β ≠ β0

H 0 : μ=μ 0 −t (n−k ) α t ←t ( n−k ) α

Page 3 of 5 Le Thanh Ha
AOF – IIFE – ECON1150 – Econometrics for Finance

H 1: β ≠ β0

Important test

H 0 : β j=0: coefficient is statistically insignificant

H 1 : β j ≠ 0 : coefficient is statistically significant

Using P−value :

Overall significant test

H 0 : β 2=…=β k =0: model is overall insignificant

H 1 : Not H 0: model is overall significant : at least….

F-test: F−stat > F (k−1 ,n−k ) α: reject H 0

Using P−value : P_value < alpha: reject H 0

DUMMY VARIABLE

Binary variable: two categories: A and Ā

Dummy: D=1 at A; D=0: otherwise

Model (1): Y t =β 1+ α Dt +u t

Model (2): Y t =β 1+ β2 X +α 1 Dt +ut

Model (3): Y t =β 1+ β2 X +α 2 D t∗X t +ut

Model (4): Y t =β 1+ β2 X +α 1 Dt +α 2 Dt∗X t +ut

Page 4 of 5 Le Thanh Ha
AOF – IIFE – ECON1150 – Econometrics for Finance

Example: Note: 4 seasons we need to create 3 dummies

Gender: Male vs Female Only create 1 dummy, e’g’, Male=1 if responder is male, 0 otherwise.

Wage = Bo+ B1*Educ+B2*Exp+B3*Male+…

Bo (intercept or constant term) is a average wage of workers.

Male is a dummy.

 Male =1  Wage = Bo+B3


 Male =0  Wage = Bo

The difference between the wage of male and female worker is B3.

Wage = Bo+ B1*Educ+B2*Exp+ B4*Educ*Male

Gender is considered as a mediator that influences the impacts of Educ on wage.

 Male =1  Wage = Bo+(B1+B4)*Educ


 Male =0  Wage = Bo+B1*Educ

Page 5 of 5 Le Thanh Ha

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy