Methods of estimation

• In simultaneous equation models (SEMs), estimation techniques

are crucial for obtaining consistent and efficient parameter
estimates.
• In an SEM, two approaches may be adopted to estimate the
structural equations, namely, single-equation methods, also
known as limited information methods, and system methods,
also known as full information methods.
• In the single-equation methods, we estimate each equation in
the system (of simultaneous equations) individually, taking into
account any restrictions placed on that equation (such as
exclusion of some variables) without worrying about the
restrictions on the other equations in the system, hence the name
limited information methods.
• In the system methods, on the other hand, we estimate all the
equations in the model simultaneously, taking due account of
all restrictions on such equations by the omission or absence
of some variables (recall that for identification such
restrictions are essential), hence the name full information
methods.
• Although the systems method-such as the full information
maximum likelihood (FIML) method- may be good to
preserve the spirit of simultaneous-equation models, in reality
they are not commonly used for different reasons:
Some of these reasons are:
 High burden of the computation: for example, estimating 20 equations
require incorporating 151 coefficients for US economy in 1955.
 The systems methods, such as FIML, lead to solutions that are highly non-
linear in the parameters and are, therefore, often difficult to determine.

 If there is a specification error (say, a wrong functional form or exclusion

of relevant variables) in one or more equations of the system, that error is
transmitted to the rest of the system. As a result, the systems methods
become very sensitive to specification errors.
 Due to the above problems, therefore, single-equation methods are often
used in practice. These are Ordinary least squares (OLS),Indirect least
squares (ILS) and Two-stage least squares (2SLS)
• INDIRECT LEAST SQUARES (ILS)
METHOD
• The method of ILS is used to obtain estimates of
the structural coefficients in a just or exactly
identified structural equation.
• These estimates are derived from the Ordinary
Least Squares (OLS) estimates of the reduced-
form coefficients.
• The estimates obtained through this process are
referred to as indirect least-squares estimates.
• For a just or exactly identified structural equation, the method of
obtaining the estimates of the structural coefficients from the OLS
estimates of the reduced-form coefficients is known as the method of
indirect least squares (ILS), and the estimates thus obtained are
known as the indirect least-squares estimates.
• The ILS method involves the following three steps:
• Step 1: Obtain the reduced-form equations. The reduced-form
equations are derived from the original structural equation by
eliminating any endogenous variables and expressing them solely in
terms of exogenous variables.
• Step 2: Apply OLS to the reduced-form equations
individually. OLS estimation is performed on each
reduced-form equation separately. This involves regressing
the endogenous variables in each equation on the
exogenous variables, using OLS to obtain the estimated
coefficients.

• Step 3: Obtain estimates of the original structural

coefficients from the estimated reduced-form
coefficients obtained in Step 2. Using the estimated
coefficients from the reduced-form equations, the original
structural coefficients are derived. This step involves a
mathematical transformation or inversion process that
allows for the estimation of the structural coefficients.
• The first reduced-form equation can be derived by substituting the
second equation into the first and rearranging:
• The second reduced-form equation can be derived by
substituting the first equation into the second and
rearranging:

𝑌𝑡 =Π2 +Π3 𝐼𝑡 +𝑣2𝑡

Where Π2 = 𝛽 0 +𝛽 1 𝛼0 , Π3 =𝛽1 𝛼1 +𝛽2 , 𝑣2𝑡 =𝛽1 𝑈1 t+𝑈2𝑡

• A unique value of the structural parameters of
the exactly identified M equation can be
calculated from the reduced-form parameters
of as follows:
Exercise: Give the following data estimate using ILS method:
 Firstly, regress Mt on i:

m Coef. Std. Err. t P>t [95% Conf. Interval]

i .1874091 .1720479 1.09 0.297 -.1874511 .5622693

_cons 58.71862 16.73577 3.51 0.004 22.2545 95.18273

The estimated result of the first reduced form equation ,Mt

​ from the above result, Π0 =5 8.71 and Π1 =0.187

 Secondly, regress Yt on i:
[95%
y Coef. Std. Err. t P>t Interval]
Conf.

i 84.14454 78.84863 1.07 0.307 -87.65187 255.941

_cons 765.5818 7669.915 0.10 0.922 -15945.73 17476.89

The estimated result of the second reduced

form equation ,Yt

​ from the above result, Π2 =84.14 and

Π3 =765.58
• Therefore, ILS results become:
Π1 0.187
• 𝛼1 = = =0.00024
Π3 765.58
• 𝛼0 =Π0 -𝛼1 Π2 = 5 8.71 –(0.00024)*(84.14
)=58.69
• Remember from chapter of econometrics I
that estimating equation( ) using OLS
results manually:

This equation can be minimized in to:

• Two stage least square
• The Two-Stage Least Squares (2SLS) estimator is
an extension of the Instrumental Variable (IV)
estimator and provides a solution for estimating
structural equations in the presence of
endogeneity.
• It is particularly useful when the equation of
interest is over identified.
• However, if the equation is exactly identified, the
2SLS estimator produces similar result to ILS.
• The 2SLS estimator involves two sequential steps, both
utilizing the Ordinary Least Squares (OLS) estimator:
• Stage 1: In this stage, each endogenous variable on the right-
hand side of the equation to be estimated is regressed on all
exogenous variables in the simultaneous equation model. This
is done using the OLS estimator, and the resulting regression
model provides fitted values for each of the endogenous
variables.
• Stage 2: In the equation to be estimated, each endogenous
right-hand side variable is replaced by its corresponding fitted
value obtained from Stage 1.
• The equation is then estimated using the OLS estimator,
treating the fitted values as if they were the original
• For equations that are exactly identified, both the
Indirect Least Squares (ILS) and 2SLS estimators
yield identical results.
• However, the 2SLS estimator is preferred because it
also provides standard errors, which are essential for
assessing the precision and statistical significance of
the estimated coefficients.