BUS 104 :INTRODUCTION TO ECONOMETRICS
INTRODUCTION TO ECONOMETRICS
Definition and scope of econometrics
It is the quantitative analysis of actual economic phenomena based on the concurrent
development of theory and observation, related by appropriate methods of inference.
It is a social science in which the tools of economic theory, mathematics and statistical
inference are applied to the analysis of economic phenomena.
It is the integration of economics, mathematics and statistics for purpose of providing
numerical values for the parameters of economic relationships and verifying economic
theories.
Relationship between economic theory, mathematics and statistics
Economic theory makes statements or hypothesis that are mostly qualitative or verbal
exposition in nature e.g. micro-economic theory postulates that a reduction in the price
of a commodity is expected to increase the quantity demanded of that commodity.
Economic theory states a negative or inverse relationship between the price and
quantity demanded but it does not provide any numerical measure of the relationship
between the two i.e. it does not tell by how much the quantity change as a result of a
certain change in the price of the commodity.
Mathematical economists express economic theory in mathematical form (equations)
without regard to measurability or empirical verification of the theory. They assume
economic relationship is exact and do not allow for random disturbances.
Economics statistics is mainly concerned with collecting, processing and presenting
economic data in the form of diagrams and tables.
Economic theory Mathematical economics Econometrics
It indicates verbal It states economic It states economic
explanation and statements relationships / theory in relationships / theory in
terms of mathematical terms of mathematical
symbols symbols
It expresses the various It expresses the various Assumes that relationships
relationships in an exact relationships in an exact are not exact ( stochastic)
form form
It does not allow for random It does not allow for random Takes into account random
element and the relationship element and the relationship disturbances
are of non- are of non-stochastic form
stochastic form
It does not provide It provides numerical values It provides numerical values
numerical values of the of the coefficients of the of the coefficients of the
coefficients of the economic phenomena economic phenomena
economic phenomena
It does not make any It does not make any It makes the empirical
Page 1 of 33
, BUS 104 :INTRODUCTION TO ECONOMETRICS
empirical verification of empirical verification of verification of economic
economic relationships / economic relationships / relationships / theory
theory theory
Model Building in Econometrics
A model consists of a number of elements abstracted from reality and some statements
about the relationships, which exist between them. An economic model is a logical (usually
mathematical) representation of whatever a prior or theoretical knowledge of economic
analysis suggests is most relevant for treating a particular problem. Usually a model takes
the form of a system of equations embodying certain assumed independencies among more
or less operationally defined variables. A model can be described by a list of endogenous
variables, a list of exogenous variables and a list of specifications as to which variables are
present in each equation. In econometric modeling process, a number of concepts and
terms are used which include variables, equations, data and graphs.
Variables
A requirement of good model construction is ensuring that all the variables an
econometrician seeks to explain are consistent with an underlying model. A variable may
be defined as a quantity that can assume any value from a given set of numbers. In
economics, a variable is a quantity or magnitude, which varies from one individual
observation to another or over a period of time under consideration. A variable can be
quantitative or qualitative. The simultaneous equation model is a complete and systematic
approach to estimation. To develop a complete model, it is necessary to describe the
variables that are used in the models. They are endogenous, exogenous variables and
lagged endogenous variables.
Endogenous variables: they are variables whose values are required to be explained
or predicted or generated by the model. They are determined within the system of
equations
Exogenous variables: they are variables whose values are not to be determined within
the system but they are assumed to be given or known or known in advance
independently.
Lagged endogenous variable: these are lagged values of the endogenous variables
e.g. demand may depend not only on current period price (Pt) but also on price of the
previous period (Pt-1). Pt is the endogenous variable in period t and Pt-1 is the lagged
endogenous variable.
Equations
An equation is a statement of equality between two mathematical arguments. An equation
contains one or more variables, the coefficients and exponents associated with these
variables, and a constant term giving the value of the equation. A set of equations
Page 2 of 33
, BUS 104 :INTRODUCTION TO ECONOMETRICS
has to be devised to express correctly the relationship between the variables. Economic
equations contain the following types:-
Behavioural equation: It describes the actions of individuals or groups or firms in the
economy. It may express a function of one independent variable or several independent
variables e.g. demand function, consumption function, investment function etc
Identities or Definitional equations: These equations assume that the relationship
between variables is true by its definition e.g. Disposable income = Consumption +
Savings
Technological equations: they are the descriptions of the production process in the
economy.
Data
Estimation of econometric model and its reliability mainly depends upon availability of
appropriate data. The data used in the estimation of a model are generally of the following
types:-
Time series: Time series data gives information about the numerical values of
variables over a period of time.
Cross sectional data: These data give information on one or more variables
collected at a given point of time.
Panel data: These are data collected from the repeated surveys of a single (cross
section) sample in different periods or points of time.
Engineering data: These data provide the information about the technical
requirements in the production process and technical relationship between output
and input requirement in a firm or an industry or the economy as a whole.
Graphical analysis
Graphs are visual aids enabling us to trace the functional relationships clearly between the
concerned economic variables and make further analysis very easy. Graphical
representations or pictures of relations between two variables are very prominent in
economics. Any economic proposition that can be specified as a function of one or two
independent variables can be illustrated with the aid of a graph. It is always good practice
to do some preliminary analysis and inspection of the data before going to further analysis.
By plotting the data or graph, one can gauge the extent of and pattern of variation. Thus,
scatter diagram would be useful to choose an appropriate functional form for further steps
in econometric modeling.
Goals of econometrics
Analysis: econometric models have been extensively used to verify the validity of
economic theories. The analysis aims at knowing the real world phenomenon by
quantifying, measuring, testing and validating economic relationship. Econometrics
Page 3 of 33
, BUS 104 :INTRODUCTION TO ECONOMETRICS
enables the researcher to evaluate existing theories and formulate new theories by
rejecting or revising the existing ones.
Policy making: estimated numerical coefficients of economic relationships help to
formulate sound economic policies and to choose the best one among the alternative
policies or compare the effects of alternative policy decisions.
Forecasting: Apart from analysis and policy making, econometrics aims at forecasting
the future values of the variable(s). the forecasts provide the basis for specific action to
the policy makers and government.
Limitations of Econometric Method
Some of the deficiencies are attributable to the inadequate specification of economic
theory itself. Most of the economic theory depends too heavily on uncontrollable
individual behavior patterns and technological constraints as the basic determinants.
The discretionary component of human and or managerial behavior and its impact on
dependent variable is inadequately represented.
Errors may arise due to wrong assumptions about the relationship among the variables
considered under the study. It is possible that economists may assume the relationship
between variables as linear, whereas it may be a non-linear one. In addition to this,
econometric models are time consuming, tedious and complex and this requires
specialized skill and knowledge on the part of the researchers.
While estimating the econometric model, it has to fulfill the assumptions laid down by
the econometric theory. If the model fails to satisfy these assumptions, many more
problems will arise.
Errors may also arise due to mis-specification of econometric models. If anyone of the
relevant variables is left out and (or) any one of the irrelevant variables is included in
the model, mis-specification error is bound to occur in the model. Too many variables
and equations in the model make the analysis difficult and complex.
Econometric methods mainly rely on quantitative data. They throw very little light on
qualitative problems and fail to explain the phenomenon. In addition, data are subject
to inaccuracy in reporting, as is common with human behavior.
Even when data limitations are not severe, the inadequacy of theoretical specification
makes the analysis meaningless. When data and theory are in conflict the researcher is
unable to arrive at meaningful conclusions.
Errors may arise due to independent variables when they move in the same direction
making it impossible to find their effects on dependent variables.
If we take cross-section data at a given point in time, then prices will be the same for
households. Hence the price effect on quantity demanded will not be determined.
Page 4 of 33
INTRODUCTION TO ECONOMETRICS
Definition and scope of econometrics
It is the quantitative analysis of actual economic phenomena based on the concurrent
development of theory and observation, related by appropriate methods of inference.
It is a social science in which the tools of economic theory, mathematics and statistical
inference are applied to the analysis of economic phenomena.
It is the integration of economics, mathematics and statistics for purpose of providing
numerical values for the parameters of economic relationships and verifying economic
theories.
Relationship between economic theory, mathematics and statistics
Economic theory makes statements or hypothesis that are mostly qualitative or verbal
exposition in nature e.g. micro-economic theory postulates that a reduction in the price
of a commodity is expected to increase the quantity demanded of that commodity.
Economic theory states a negative or inverse relationship between the price and
quantity demanded but it does not provide any numerical measure of the relationship
between the two i.e. it does not tell by how much the quantity change as a result of a
certain change in the price of the commodity.
Mathematical economists express economic theory in mathematical form (equations)
without regard to measurability or empirical verification of the theory. They assume
economic relationship is exact and do not allow for random disturbances.
Economics statistics is mainly concerned with collecting, processing and presenting
economic data in the form of diagrams and tables.
Economic theory Mathematical economics Econometrics
It indicates verbal It states economic It states economic
explanation and statements relationships / theory in relationships / theory in
terms of mathematical terms of mathematical
symbols symbols
It expresses the various It expresses the various Assumes that relationships
relationships in an exact relationships in an exact are not exact ( stochastic)
form form
It does not allow for random It does not allow for random Takes into account random
element and the relationship element and the relationship disturbances
are of non- are of non-stochastic form
stochastic form
It does not provide It provides numerical values It provides numerical values
numerical values of the of the coefficients of the of the coefficients of the
coefficients of the economic phenomena economic phenomena
economic phenomena
It does not make any It does not make any It makes the empirical
Page 1 of 33
, BUS 104 :INTRODUCTION TO ECONOMETRICS
empirical verification of empirical verification of verification of economic
economic relationships / economic relationships / relationships / theory
theory theory
Model Building in Econometrics
A model consists of a number of elements abstracted from reality and some statements
about the relationships, which exist between them. An economic model is a logical (usually
mathematical) representation of whatever a prior or theoretical knowledge of economic
analysis suggests is most relevant for treating a particular problem. Usually a model takes
the form of a system of equations embodying certain assumed independencies among more
or less operationally defined variables. A model can be described by a list of endogenous
variables, a list of exogenous variables and a list of specifications as to which variables are
present in each equation. In econometric modeling process, a number of concepts and
terms are used which include variables, equations, data and graphs.
Variables
A requirement of good model construction is ensuring that all the variables an
econometrician seeks to explain are consistent with an underlying model. A variable may
be defined as a quantity that can assume any value from a given set of numbers. In
economics, a variable is a quantity or magnitude, which varies from one individual
observation to another or over a period of time under consideration. A variable can be
quantitative or qualitative. The simultaneous equation model is a complete and systematic
approach to estimation. To develop a complete model, it is necessary to describe the
variables that are used in the models. They are endogenous, exogenous variables and
lagged endogenous variables.
Endogenous variables: they are variables whose values are required to be explained
or predicted or generated by the model. They are determined within the system of
equations
Exogenous variables: they are variables whose values are not to be determined within
the system but they are assumed to be given or known or known in advance
independently.
Lagged endogenous variable: these are lagged values of the endogenous variables
e.g. demand may depend not only on current period price (Pt) but also on price of the
previous period (Pt-1). Pt is the endogenous variable in period t and Pt-1 is the lagged
endogenous variable.
Equations
An equation is a statement of equality between two mathematical arguments. An equation
contains one or more variables, the coefficients and exponents associated with these
variables, and a constant term giving the value of the equation. A set of equations
Page 2 of 33
, BUS 104 :INTRODUCTION TO ECONOMETRICS
has to be devised to express correctly the relationship between the variables. Economic
equations contain the following types:-
Behavioural equation: It describes the actions of individuals or groups or firms in the
economy. It may express a function of one independent variable or several independent
variables e.g. demand function, consumption function, investment function etc
Identities or Definitional equations: These equations assume that the relationship
between variables is true by its definition e.g. Disposable income = Consumption +
Savings
Technological equations: they are the descriptions of the production process in the
economy.
Data
Estimation of econometric model and its reliability mainly depends upon availability of
appropriate data. The data used in the estimation of a model are generally of the following
types:-
Time series: Time series data gives information about the numerical values of
variables over a period of time.
Cross sectional data: These data give information on one or more variables
collected at a given point of time.
Panel data: These are data collected from the repeated surveys of a single (cross
section) sample in different periods or points of time.
Engineering data: These data provide the information about the technical
requirements in the production process and technical relationship between output
and input requirement in a firm or an industry or the economy as a whole.
Graphical analysis
Graphs are visual aids enabling us to trace the functional relationships clearly between the
concerned economic variables and make further analysis very easy. Graphical
representations or pictures of relations between two variables are very prominent in
economics. Any economic proposition that can be specified as a function of one or two
independent variables can be illustrated with the aid of a graph. It is always good practice
to do some preliminary analysis and inspection of the data before going to further analysis.
By plotting the data or graph, one can gauge the extent of and pattern of variation. Thus,
scatter diagram would be useful to choose an appropriate functional form for further steps
in econometric modeling.
Goals of econometrics
Analysis: econometric models have been extensively used to verify the validity of
economic theories. The analysis aims at knowing the real world phenomenon by
quantifying, measuring, testing and validating economic relationship. Econometrics
Page 3 of 33
, BUS 104 :INTRODUCTION TO ECONOMETRICS
enables the researcher to evaluate existing theories and formulate new theories by
rejecting or revising the existing ones.
Policy making: estimated numerical coefficients of economic relationships help to
formulate sound economic policies and to choose the best one among the alternative
policies or compare the effects of alternative policy decisions.
Forecasting: Apart from analysis and policy making, econometrics aims at forecasting
the future values of the variable(s). the forecasts provide the basis for specific action to
the policy makers and government.
Limitations of Econometric Method
Some of the deficiencies are attributable to the inadequate specification of economic
theory itself. Most of the economic theory depends too heavily on uncontrollable
individual behavior patterns and technological constraints as the basic determinants.
The discretionary component of human and or managerial behavior and its impact on
dependent variable is inadequately represented.
Errors may arise due to wrong assumptions about the relationship among the variables
considered under the study. It is possible that economists may assume the relationship
between variables as linear, whereas it may be a non-linear one. In addition to this,
econometric models are time consuming, tedious and complex and this requires
specialized skill and knowledge on the part of the researchers.
While estimating the econometric model, it has to fulfill the assumptions laid down by
the econometric theory. If the model fails to satisfy these assumptions, many more
problems will arise.
Errors may also arise due to mis-specification of econometric models. If anyone of the
relevant variables is left out and (or) any one of the irrelevant variables is included in
the model, mis-specification error is bound to occur in the model. Too many variables
and equations in the model make the analysis difficult and complex.
Econometric methods mainly rely on quantitative data. They throw very little light on
qualitative problems and fail to explain the phenomenon. In addition, data are subject
to inaccuracy in reporting, as is common with human behavior.
Even when data limitations are not severe, the inadequacy of theoretical specification
makes the analysis meaningless. When data and theory are in conflict the researcher is
unable to arrive at meaningful conclusions.
Errors may arise due to independent variables when they move in the same direction
making it impossible to find their effects on dependent variables.
If we take cross-section data at a given point in time, then prices will be the same for
households. Hence the price effect on quantity demanded will not be determined.
Page 4 of 33
