ECON20003 Lecture Notes - Lecture 10: Sampling Distribution, Statistical Hypothesis Testing, Heteroscedasticity

17 views2 pages
Liea assoiatio etee to aiales a e epeseted a lie, fitted least suaes
Fitted least-suaes lie itte as
=
+
Paaete estiates ae alled OLS estiates fo udelig populatio paaetes
Last Tie
The diffeee etee oseed Yi ad its oespodig fitted alue
Y
 is alled the esidual:
Rado saplig assuptio
ε
i
ad X
i
ae idepedet ad E
ε
i
=
Hoosedastiit
aiae of
ε
is a ostat
At least to alues of X ust e diffeet
Assuptios of the Siple Regessio Model
Assuptio
Rado saple da fo a populatio hih satisfies the elatioship:
Y
i
is outoe o depedet aiale
X
i
is the eplaato o idepedet aiale o egesso
ε
i
is the eo te o distuae
Betas ae paaetes e at to estiate
Iplies pais Y, X ae idepedet
Y is ot idepedet of X
the'e lieal elated
Hoee, Y
, X
is idepedet of Y
, X
Assuptio iolated fo tie
-
seies oseatios
ol oks fo oss
-
setioal data
RANDOM SAMPLING ASSUMPTION
Assuptio
ε
i
ad X
i
ae idepedet ad E
ε
i
=
Coditioal ea futio fo Y
Itepetatio of oeffiiet
β
To ipliatios of this assuptio:
MEAN FOR Y IS CONDITIONAL ON X
Vaiale Y ill still hae a oetioal uoditioal ea
Coditioal ea EY/X efes to the ea of Y fo a suset of the populatio ith a speifi alue fo X
Coditioig aiale is teated as fied o o
-
ado, thus:
Whe X ad
ε
ae idepedet:
Regessio futio
eoes:
If assuptio holds a itepet
β
as the "hage i Y
aused
a hage i X"
If
ε
ad X ae oelated e a't gie
β
a ausal itepetatio
Istead, a sa that
β
is the "hage i Y
assoiated
ith a hage i X"
ε
eflets otiutio of all aiales, othe tha X, that ipat Y
Hae
ε
ad X idepedet eas the'e uoelated, hee:
INTERPRETATIONS OF COEFFICIENTS
Fo the egessio euatio
=
+
:
Lecture : Siple Regressio
Thusda,  August 7 : AM
Linear Regression Page 1
find more resources at oneclass.com
find more resources at oneclass.com
Unlock document

This preview shows half of the first page of the document.
Unlock all 2 pages and 3 million more documents.

Already have an account? Log in

Document Summary

Li(cid:374)ea(cid:396) asso(cid:272)iatio(cid:374) (cid:271)et(cid:449)ee(cid:374) t(cid:449)o (cid:448)a(cid:396)ia(cid:271)les (cid:272)a(cid:374) (cid:271)e (cid:396)ep(cid:396)ese(cid:374)ted (cid:271)(cid:455) a li(cid:374)e, fitted (cid:271)(cid:455) least s(cid:395)ua(cid:396)es. The diffe(cid:396)e(cid:374)(cid:272)e (cid:271)et(cid:449)ee(cid:374) o(cid:271)se(cid:396)(cid:448)ed (cid:894)yi(cid:895) a(cid:374)d its (cid:272)o(cid:396)(cid:396)espo(cid:374)di(cid:374)g fitted (cid:448)alue (cid:894)y(cid:3553)(cid:2919)(cid:895) is (cid:272)alled the (cid:396)esidual: Pa(cid:396)a(cid:373)ete(cid:396) esti(cid:373)ates a(cid:396)e (cid:272)alled ols esti(cid:373)ates fo(cid:396) u(cid:374)de(cid:396)l(cid:455)i(cid:374)g populatio(cid:374) pa(cid:396)a(cid:373)ete(cid:396)s. Assu(cid:373)ptio(cid:374)s of the si(cid:373)ple reg(cid:396)essio(cid:374) model (cid:1005)(cid:895) (cid:1006)(cid:895) (cid:1007)(cid:895) (cid:1008)(cid:895) Ho(cid:373)os(cid:272)edasti(cid:272)it(cid:455) (cid:448)a(cid:396)ia(cid:374)(cid:272)e of is a (cid:272)o(cid:374)sta(cid:374)t. At least t(cid:449)o (cid:448)alues of x (cid:373)ust (cid:271)e diffe(cid:396)e(cid:374)t. Ra(cid:374)do(cid:373) sa(cid:373)ple d(cid:396)a(cid:449)(cid:374) f(cid:396)o(cid:373) a populatio(cid:374) (cid:449)hi(cid:272)h satisfies the (cid:396)elatio(cid:374)ship: Xi is the e(cid:454)pla(cid:374)ato(cid:396)(cid:455) o(cid:396) i(cid:374)depe(cid:374)de(cid:374)t (cid:448)a(cid:396)ia(cid:271)le (cid:894)o(cid:396) (cid:396)eg(cid:396)esso(cid:396)(cid:895) Ho(cid:449)e(cid:448)e(cid:396), (cid:894)y(cid:1005), x(cid:1005)(cid:895) is i(cid:374)depe(cid:374)de(cid:374)t of (cid:894)y(cid:1006), x(cid:1006)(cid:895) Y is (cid:374)ot i(cid:374)depe(cid:374)de(cid:374)t of x the(cid:455)"(cid:396)e li(cid:374)ea(cid:396)l(cid:455) (cid:396)elated. Assu(cid:373)ptio(cid:374) (cid:448)iolated fo(cid:396) ti(cid:373)e-se(cid:396)ies o(cid:271)se(cid:396)(cid:448)atio(cid:374)s o(cid:374)l(cid:455) (cid:449)o(cid:396)ks fo(cid:396) (cid:272)(cid:396)oss-se(cid:272)tio(cid:374)al data. I a(cid:374)d xi a(cid:396)e i(cid:374)depe(cid:374)de(cid:374)t a(cid:374)d e(cid:894) i(cid:895) = (cid:1004) Va(cid:396)ia(cid:271)le y (cid:449)ill still ha(cid:448)e a (cid:272)o(cid:374)(cid:448)e(cid:374)tio(cid:374)al u(cid:374)(cid:272)o(cid:374)ditio(cid:374)al (cid:373)ea(cid:374) Co(cid:374)ditio(cid:374)al (cid:373)ea(cid:374) e(cid:894)y/x(cid:895) (cid:396)efe(cid:396)s to the (cid:373)ea(cid:374) of y fo(cid:396) a su(cid:271)set of the populatio(cid:374) (cid:449)ith a spe(cid:272)ifi(cid:272) (cid:448)alue fo(cid:396) x. Co(cid:374)ditio(cid:374)i(cid:374)g (cid:448)a(cid:396)ia(cid:271)le is t(cid:396)eated as fi(cid:454)ed o(cid:396) (cid:374)o(cid:374)-(cid:396)a(cid:374)do(cid:373), thus:

Get access

Grade+20% off
$8 USD/m$10 USD/m
Billed $96 USD annually
Grade+
Homework Help
Study Guides
Textbook Solutions
Class Notes
Textbook Notes
Booster Class
40 Verified Answers
Class+
$8 USD/m
Billed $96 USD annually
Class+
Homework Help
Study Guides
Textbook Solutions
Class Notes
Textbook Notes
Booster Class
30 Verified Answers

Related Documents