Scribd Logo
Upload

Welcome to Scribd!

  • 23 views

    1 C-D Problem: 1.1 Unknown

    Uploaded by

    Eyup Dogan
    This document contains lecture notes on econometrics concepts including: (1) The Cobb-Douglas production function model and derivation of the conditional expectation and firms' profit maximization conditions. (2) Estimation of the coefficient β using the ordinary least squares method. (3) The maximum likelihood estimation approach assuming errors are normally distributed and independently and identically distributed. (4) A mathematical proof showing that the condition that the conditional expectation of μ given x is equal to 0 implies they are uncorrelated, but the reverse is not necessarily true, using an example of a counter case.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    1 C-D Problem: 1.1 Unknown

    Uploaded by

    Eyup Dogan
    23 views4 pages
    This document contains lecture notes on econometrics concepts including: (1) The Cobb-Douglas production function model and derivation of the conditional expectation and firms' profit maximization conditions. (2) Estimation of the coefficient β using the ordinary least squares method. (3) The maximum likelihood estimation approach assuming errors are normally distributed and independently and identically distributed. (4) A mathematical proof showing that the condition that the conditional expectation of μ given x is equal to 0 implies they are uncorrelated, but the reverse is not necessarily true, using an example of a counter case.

    Original Title

    sol_HW1_Di

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains lecture notes on econometrics concepts including: (1) The Cobb-Douglas production function model and derivation of the conditional expectation and firms' profit maximization conditions. (2) Estimation of the coefficient β using the ordinary least squares method. (3) The maximum likelihood estimation approach assuming errors are normally distributed and independently and identically distributed. (4) A mathematical proof showing that the condition that the conditional expectation of μ given x is equal to 0 implies they are uncorrelated, but the reverse is not necessarily true, using an example of a counter case.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    23 views4 pages

    1 C-D Problem: 1.1 Unknown

    Uploaded by

    Eyup Dogan
    This document contains lecture notes on econometrics concepts including: (1) The Cobb-Douglas production function model and derivation of the conditional expectation and firms' profit maximization conditions. (2) Estimation of the coefficient β using the ordinary least squares method. (3) The maximum likelihood estimation approach assuming errors are normally distributed and independently and identically distributed. (4) A mathematical proof showing that the condition that the conditional expectation of μ given x is equal to 0 implies they are uncorrelated, but the reverse is not necessarily true, using an example of a counter case.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 4

    Econ 807 Spring 2012 Prof.

    Mroz Solution for HW I provided by Di Wu

    1
    1.1

    C-D problem
    Unknown
    max E() = E(P yi rki ) = E(P AKi e i rki )
    1 FOC:P Aki E(e ) r = 0

    (1) (2) (3) (4)

    ki = [ Notice, suppose E(e i ) = e+


    2 2

    1 r ] 1 P AE(e )

    N (0, 2 ), we know that E( i ) = 0, however, e i lognormal, and


    2 2

    =e

    =0

    1.2

    Firms know
    max = P yi rki = P AKi e i rki
    1 FOC:P Aki e r = 0

    (5) (6) (7) (8)

    ki = [

    1 r ] 1 P Ae

    Estimation of
    b = = = = = (xi x)(yi y ) 2 (xi x) (xi yi xi y xy + xy ) i 2 (xi 2xi x + x2 ) (xi yi ) ny ny + ny x x x 2 2 + n2 (xi ) 2n x x 2 n (xi yi ) n xy 2 2 x2 n (xi ) n n (xi yi ) xi yi n (x2 ) ( xi )2 i (9) (10) (11) (12) (13)

    MLE
    i

    Assuming

    N (0, 2 ), and

    are i.i.d. random variables.


    ( i )2 1 e 22 2

    LF =
    i

    (14)

    Taking the log of last equation wont change the value of which the likelihood function achieves maximum due to the increasing of log transformation, max, LLF =
    i

    (yi xi )2 ) (log( 2) 2 2

    (15)

    Since (log( 2) includes no information on and , so the previous maximizition problem is identical to the following, max
    i

    (yi xi )2

    (16)

    which is identical to OLS, which is, min


    i

    (yi xi )2

    (17)

    Which condition is stronger?

    To show this mathematically, you need to know law of iterated expectation, which is, E(E(x|y)) = E(x) 2 (18)

    proof: E(E(x|y)) = = = fY (y) fY (y) xf (x|y)dxdy x f (x, y) dxdy f (y) (19) (20) (21) (22)

    fY (y) fY (y) = E(x)

    xf (x, y)dxdy

    Now remind the logic I used to show that is show I: E(|x) = 0 is implying 0 covariance, II: 0 covariance is identical to E(x) = 0, III: 0 covariance can not imply E(|x) = 0. Proof of I: rst we knew that E(|x) = 0 implies E(|x) = E() = 0 E(|x) = = = f (|x)d f (x, ) d f (x) f (x, )d (23) (24) (25) (26)

    1 f (x) = 0 so we have

    f (x, )d = 0, which will imply that E(|x) = E() = 0. E(|x) = E( = x |x) x (27) (28)

    1 E(x|x) x

    which implies, xE(|x) = E(x|x) Taking expectaion on both side, E(E(x|x)) = E(x) = E(xE(|x)) = E(xE()) = E(x)E() (30) (31) (32) (33) (29)

    By the denition of cov(x, ) = E(x) E(x)E(), from last equation we know that assuming E(|x) = 0, we have 0 covariance, so we have showed I. 3

    Since we know that we could assume E() = 0, and cov(x, ) = E(x) E(x)E() = 0, then E(x) = 0. So we have shown II. The only last thing is III. We show this by nd a counter example. Assuming E(|x) = x2 , and x N (0, 1). So we know that E(x) = 0, E() = E(E(|x)) = E(x2 ) = 1, since x2 2 (1). E(x) = E(xE(|x)) = E(x3 ) = 0, since skewness of normal distribution is 0. So, cov(x, ) = E(x) E(x)E() = 0 0 = 0. However, E(|x) = x2 = 0. We show all 3 parts.

    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