slides_Ch2_W[1]

G a person drive s melissa tartari yale econometrics 4

Info iconThis preview shows page 1. Sign up to view the full content.

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

Unformatted text preview: y be factors that are: inherently unobservable/unmeasurable (e.g. a person’ drive); s Melissa Tartari (Yale) Econometrics 4 / 93 De…nition: the SLRM III U represents factors other than X that a¤ect Y ; we can think of U as standing, loosely, for “unobserved”; these may be factors that are: inherently unobservable/unmeasurable (e.g. a person’ drive); s di¢ cult or very expensive to observe/measure (e.g. a person’ time s diary); Melissa Tartari (Yale) Econometrics 4 / 93 De…nition: the SLRM III U represents factors other than X that a¤ect Y ; we can think of U as standing, loosely, for “unobserved”; these may be factors that are: inherently unobservable/unmeasurable (e.g. a person’ drive); s di¢ cult or very expensive to observe/measure (e.g. a person’ time s diary); observable and/or observed but “not the focus of the analysis”. Melissa Tartari (Yale) Econometrics 4 / 93 De…nition: the SLRM III U represents factors other than X that a¤ect Y ; we can think of U as standing, loosely, for “unobserved”; these may be factors that are: inherently unobservable/unmeasurable (e.g. a person’ drive); s di¢ cult or very expensive to observe/measure (e.g. a person’ time s diary); observable and/or observed but “not the focus of the analysis”. Wlog we normalize the unconditional expectation of u to zero: E [U ] = 0 Melissa Tartari (Yale) Econometrics (2.5) 4 / 93 De…nition: the SLRM III U represents factors other than X that a¤ect Y ; we can think of U as standing, loosely, for “unobserved”; these may be factors that are: inherently unobservable/unmeasurable (e.g. a person’ drive); s di¢ cult or very expensive to observe/measure (e.g. a person’ time s diary); observable and/or observed but “not the focus of the analysis”. Wlog we normalize the unconditional expectation of u to zero: E [U ] = 0 (2.5) Why wlog? Suppose not. Let E [U ] = λ 6= 0 then E [Y ] = ( βo + λ) + β1 E [X ] which means that we cannot separately identify βo from λ so we may as well normalize one of the two to zero, we choose to normalize λ, instead of βo , to zero for convenience. Melissa Tartari (Yale) Econometrics 4 / 93 De…nition: the SLRM III U represents factors...
View Full Document

This note was uploaded on 02/13/2014 for the course ECON 350 taught by Professor Donaldbrown during the Fall '10 term at Yale.

Ask a homework question - tutors are online