{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Econometrics-I-22

Part 22 semi and nonparametric estimation a

Info iconThis preview shows pages 14–23. Sign up to view the full content.

View Full Document Right Arrow Icon
Part 22: Semi- and Nonparametric Estimation A Nonparametric Regression p y = µ(x) +ε p Smoothing methods to approximate µ(x) at specific points, x* p For a particular x*, µ(x*) = ∑i wi(x*| x )yi n E.g., for ols, µ(x*) =a+bx* n wi = 1/n + p We look for weighting scheme, local differences in relationship. OLS assumes a fixed slope, b. ™    13/26 2 ( ) / ( ) i i i - Σ - x x x x
Background image of page 14

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Part 22: Semi- and Nonparametric Estimation Nearest Neighbor Approach p Define a neighborhood of x*. Points near get high weight, points far away get a small or zero weight p Bandwidth, h defines the neighborhood: e.g., Silverman h =.9Min[s,(IQR/1.349)]/n.2 Neighborhood is + or – h/2 p LOWESS weighting function: (tricube) Ti = [1 – [Abs(xi – x*)/h]3]3. p Weight is wi = 1[Abs(xi – x*)/h < .5] * Ti . ™    14/26
Background image of page 15
Part 22: Semi- and Nonparametric Estimation LOWESS Regression ™    15/26
Background image of page 16

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Part 22: Semi- and Nonparametric Estimation OLS Vs. Lowess ™    16/26
Background image of page 17
Part 22: Semi- and Nonparametric Estimation Smooth Function: Kernel Regression ™    17/26 1 1 2 * 1 ˆ( *| , ) * 1 Kernel Functions: Normal: K(t) = (t) Logistic: K(t) = (t)[1- (t)] Epanechnikov: K(t)=.75(1-.2t )/ 5, if |t| 5 and 0 otherwise n i i i n i i x x K y B B x B x x K B B = = - μ = - φ Λ Λ x
Background image of page 18

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Part 22: Semi- and Nonparametric Estimation Kernel Regression vs. Lowess (Lwage vs. Educ) ™    18/26
Background image of page 19
Part 22: Semi- and Nonparametric Estimation Locally Linear Regression ™    19/26 1 1 1 ( *) ( *)' *. ( *) ( *, ) ( *, ) y ( *, ) [( * ) ( * ), ] n n i i i i i i i i i i i i i i w w w K h - = = μ = β β = = - - x x x x x x x x x x x x x x x x x
Background image of page 20

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Part 22: Semi- and Nonparametric Estimation OLS vs. LOWESS ™    20/26
Background image of page 21
Part 22: Semi- and Nonparametric Estimation Quantile Regression p Least squares based on: E[y| x ]= ’x p LAD based on: Median[y| x ]= (.5)’ x p Quantile regression: Q(y| x ,q)= (q)’ x p Does this just shift the constant?
Background image of page 22

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 23
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}