Scribd
Upload
25 views14 pages

Econ T1 Eng

This document discusses key concepts in econometrics. It introduces econometrics as the development of statistical methods for estimating economic relationships and evaluating policy. It outlines different types of data used, including cross-sectional data, panel data, pooled cross-sectional data, and time series data. It also discusses the importance of accounting for causality and ceteris paribus conditions when estimating economic models empirically. Examples are provided around estimating returns to education.

Uploaded by

Enric Masclans Planas
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
25 views14 pages

Econ T1 Eng

This document discusses key concepts in econometrics. It introduces econometrics as the development of statistical methods for estimating economic relationships and evaluating policy. It outlines different types of data used, including cross-sectional data, panel data, pooled cross-sectional data, and time series data. It also discusses the importance of accounting for causality and ceteris paribus conditions when estimating economic models empirically. Examples are provided around estimating returns to education.

Uploaded by

Enric Masclans Planas
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Econometrics

Introduction

Dr. Toni Mora


Outline

l What is Econometrics and how useful is it?

l What types of data are used in (applied) econometrics?


– Cross sections
– Pooled cross sections
– Panel or longitudinal data
– Time series

l The question of causality and the notion of ceteris paribus

2
What is Econometrics?

l Econometrics is based upon the development of statistical


methods for estimating economic relationships, testing
economic theories and evaluating and implementing
government and business policy

l It has evolved as a separate discipline from mathematical


statistics because it focuses on the problem inherent in
collecting and analyzing non-experimental economic data

l Microeconometrics vs. Macroeconometrics

3
Why study Econometrics?

l Econometrics can be applied to real-world problems through


empirical analyses. In such analyses, econometric methods are
implemented using a data set to test an economic theory or to
estimate a certain relationship of interest

l Economic theory may be ambiguous as to the effect of some policy


or program change, but it is possible to use an econometric method
to evaluate such a change

l Example: What is the (net) effect of an increase in wages on labor


supply? What is the impact on salaries for any additional schooling
year?
4
Theory vs empirics

l Data measurement errors or not related to the model

l Variables not suitable to be measured (expectations)

l Theory suggests but we choose functional form

l Errors related to the stochastic properties of the random term

l Omitted variables relevant for the model


Types of Data – Cross Sectional

l Each observation is a different individual, firm, etc. with


information at one point in time

l Often cross-sectional data is a random sample (i.e. obtained


by random sampling from an underlying population)

l If the data is not a random sample, there is a sample-


selection problem

6
Pooled cross sections vs. Panel

l When data on various random cross sections from the same


underlying population are collected at various points in time,
it is possible to pool them and treat them similar to a single
cross section, after accounting for time differences
l When the same random individual observations are followed
over time, one can make explicit use of this additional
information – this is known as panel data or longitudinal data
l Panel data, while being more costly and difficult to obtain,
has several advantages over cross-sectional data or even
pooled cross-sectional data
8
Time Series

l A time series data set consists of observations on a variable or


several variables over time – e.g. stock prices (Mon-Fri),
inflation (monthly), GDP (quarterly). The time dimension is
usually very long

l A key feature of time series data is that economic observations


can rarely, if ever, be assumed to be independent across time

l One has to account for trends and seasonality effects, and other
problems such as high persistence and dynamics
10
11
% Ecologic sales evolution

.3 .4 .5 .6

3Mar19

4Mar19
Time Series

5Mar19

1Apr19

2Apr19

3Apr19

Below 35
4Apr19

5Apr

3Mar20

Bw 35-55
4Mar20

Time period (weeks 2019-2020)


5Mar20

1Apr20

2Apr20
Above 55

3Apr20

4Apr20

5Apr20
The Question of Causality

l Simply finding an association or correlation between variables is rarely


sufficient
l One wants the estimated effect of an independent variable (X) on the
dependent variable (y) to be causal
l The notion of ceteris paribus—which means “other (relevant) factors being
equal”—plays an important role in causal analysis. In analyzing consumer
demand, we are interested in knowing the effect of changing the price of a
good on its quantity demanded, while holding fixed all other factors -such
as income, prices of other goods, and individual tastes
l Except in very special cases, it will not be possible to literally hold all else
equal. The key question in most empirical studies is: Have enough other
factors been held fixed to make a case for causality?
12
Example: Returns to Education

l A simple model of human capital investment implies


that getting more education should lead to higher
earnings (Mincer model)

l In the simplest case (specifying a simple linear


regression model), this implies an equation like:

wage = β 0 + β1educ + u

13
Example (continued)

l The estimate of β1 is the return to education, but can it be


considered to be causal?
l While the error term, u, includes other factors affecting
earnings, one wants to control for as much as possible when
estimating the earlier model. Think of a model like:
wage = β 0 + β1educ + β 2 exper + u
l Still, many relevant explanatory variables remain often
unobserved to the econometrician (think of an individual’s
ability), which can be problematic if the true model is:
wage = β 0 + β1educ + β 2 exper + β 3ability + u
14

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