Differences between revisions 3 and 19 (spanning 16 versions)

Ordinary Least Squares

Ordinary Least Squares (OLS) is a linear regression method. It minimizes root mean square errors.

Contents

Ordinary Least Squares

Univariate

Given one independent variable and one dependent (outcome) variable, the OLS model is specified as:

It is estimated as:

This model describes (1) the mean observation and (2) the marginal changes to the outcome per unit changes in the independent variable.

The proof can be seen here.

Multivariate

Given k independent variables, the OLS model is specified as:

It is estimated as:

Estimated Coefficients

If these assumptions can be made:

Linearity
Exogeneity
Random sampling
No perfect multicolinearity
Homoskedasticity

Then OLS is the best linear unbiased estimator (BLUE) for regression coefficients.

The variances for each coefficient are:

Note that the standard deviation of the population's parameter is unknown, so it's estimated like:

If the homoskedasticity assumption does not hold, then the estimators for each coefficient are actually:

Wherein, for example, r_1j is the residual from regressing x₁ onto x₂, ... x_k.

The variances for each coefficient can be estimated with the Eicker-White formula:

See Nicolai Kuminoff's video lectures for the derivation of the robust estimators.

CategoryRicottone

Statistics/OrdinaryLeastSquares (last edited 2025-09-03 02:08:40 by DominicRicottone)

-  ⇤ ← Revision 3 as of 2023-10-28 05:37:48 → 
  Size: 1349
  Editor: DominicRicottone
  Comment:
+   ← Revision 19 as of 2024-06-05 22:01:56 → ⇥
  Size: 1860
  Editor: DominicRicottone
  Comment: Rewrite 2
-Deletions are marked like this.
+Additions are marked like this.
 Line 1:
-## page was renamed from Econometrics/LinearRegression
-Line 14:
+Line 13:
-The regression line passes through two points:
+Given one independent variable and one dependent (outcome) variable, the OLS model is specified as:
-Line 16:
+Line 15:
-{{attachment:regression1.svg}}
+{{attachment:model.svg}}
-Line 18:
+Line 17:
-and
+It is estimated as:
-Line 20:
+Line 19:
-{{attachment:regression2.svg}}
+{{attachment:estimate.svg}}
-Line 22:
+Line 21:
-Take the generic equation form of a line:
+This model describes (1) the mean observation and (2) the marginal changes to the outcome per unit changes in the independent variable.
-Line 24:
+Line 23:
-{{attachment:b01.svg}}
+The proof can be seen [[Econometrics/OrdinaryLeastSquares/UnivariateProof|here]].
-Line 26:
+Line 25:
-Insert the first point into this form.
+----
-Line 28:
+Line 27:
-{{attachment:b02.svg}}
-Line 30:
+Line 28:
-This can be trivially rewritten to solve for ''a'' in terms of ''b'':
-Line 32:
+Line 29:
-{{attachment:b03.svg}}
+== Multivariate ==
-Line 34:
+Line 31:
-Insert the second point into the original form.
+Given ''k'' independent variables, the OLS model is specified as:
-Line 36:
+Line 33:
-{{attachment:b04.svg}}
+{{attachment:mmodel.svg}}
-Line 38:
+Line 35:
-Now additionally insert the solution for ''a'' in terms of ''b''.
+It is estimated as:
-Line 40:
+Line 37:
-{{attachment:b05.svg}}
+{{attachment:mestimate.svg}}
-Line 42:
+Line 39:
-Expand all terms to produce:
+----
-Line 44:
+Line 41:
-{{attachment:b06.svg}}
-Line 46:
+Line 42:
-This can now be eliminated into:
-Line 48:
+Line 43:
-{{attachment:b07.svg}}
+== Estimated Coefficients ==
-Line 50:
+Line 45:
-Giving a solution for ''b'':
+If these assumptions can be made:
-Line 52:
+Line 47:
-{{attachment:b08.svg}}
+. Linearity
 2. [[Econometrics/Exogeneity|Exogeneity]]
 3. Random sampling
 4. No perfect multicolinearity
 5. [[Econometrics/Homoskedasticity|Homoskedasticity]]
-Line 54:
+Line 53:
-This solution is trivially rewritten as:
+Then OLS is the best linear unbiased estimator ('''BLUE''') for regression coefficients.
-Line 56:
+Line 55:
-{{attachment:b09.svg}}
+The variances for each coefficient are:
-Line 58:
+Line 57:
-Expand the formula for correlation as:
+{{attachment:homo1.svg}}
-Line 60:
+Line 59:
-{{attachment:b10.svg}}
+Note that the standard deviation of the population's parameter is unknown, so it's estimated like:
-Line 62:
+Line 61:
-This can now be eliminated into:
+{{attachment:homo2.svg}}
-Line 64:
+Line 63:
-{{attachment:b11.svg}}
+If the homoskedasticity assumption does not hold, then the estimators for each coefficient are actually:
-Line 66:
+Line 65:
-Finally, ''b'' can be eloquently written as:
+{{attachment:hetero1.svg}}
-Line 68:
+Line 67:
-{{attachment:b12.svg}}
+Wherein, for example, ''r,,1j,,'' is the residual from regressing ''x,,1,,'' onto ''x,,2,,'', ... ''x,,k,,''.
-Line 70:
+Line 69:
-Giving a generic formula for the regression line:
+The variances for each coefficient can be estimated with the Eicker-White formula:
-Line 72:
+Line 71:
-{{attachment:b13.svg}}
+{{attachment:hetero2.svg}}

See [[https://www.youtube.com/@kuminoff|Nicolai Kuminoff's]] video lectures for the derivation of the robust estimators.

Diff for "Statistics/OrdinaryLeastSquares"

Ordinary Least Squares

Univariate

Multivariate

Estimated Coefficients