Addi

sAbabaUni
ver
sit
y
Col
l
egeofBusi
nessandeconomi
cs
Depar
tmentofeconomi
cs
Wor
ksheetf
oreconomet
ri
csf
orf
inance
NB:DearEveryonepleasedoi
tonebyone.Yourunderst
anding
abouttheprevi
oust opi
cswil
lbeev
aluat
edbasedont hi
s
worksheet.Feelfr
eei
fyouhaveanyquest
ions.
1. Economet r
ics deal
s wi
ththe measur
ementofeconomicr el
ati
onshi
ps which ar
e
stochasti
corr andom.Thesimplestf
orm ofeconomi
crelati
onshi
psbetweent wo
vari
ablesXandYcanber epr
esent
edby:
Yi01XiUi ;where 0 1arer
egressi
onparamet
ersand
and
Uit
hest
ochast
icdi
stur
bancet
erm

Whatar
ether
easonsf
ort
hei
nser
ti
onofU-
ter
mint
hemodel
?

2. Thefoll
owi
ngdatar ef
erst
ot hedemandf
ormoney(
M)andt
herat
eofint
erest(
R)i
n
forei
ghtdi
ff
erenteconomics:
M(Inbil
li
ons) 56 50 46 30 20 35 37 61
R% 6.
3 4.
6 5.
1 7.
3 8.
9 5.
3 6.
7 3.
5

a.Assuming a M RUi,obt
aint
heOLSest
imat
orsof
rel
ati
onshi
p

and
b.Cal
cul
atet
hecoef
fi
cientofdet
ermi
nat
ionf
ort
hedat
aandi
nter
preti
tsv
alue
c. I
fina9theconomytherat
eofi
nter
esti
sR=8.
1,pr
edi
ctt
hedemandf
or
money(
M)i nthi
seconomy.
3. Thefoll
owingdat
aref
erst
othepri
ceofagood‘
P'andt
hequant
it
yoft
hegood
suppl
ied,‘
S'.
P 2 7 5 1 4 8 2 8
S 15 41 32 9 28 43 17 40
a.Est
imat
ethel
i
nearr
egr
essi
onl
i S)P
ne(

b.Est
imat
ethest
andar
der
r sofˆa
or ndˆ
c. Testt
hehy
pot
hesi
sthatpr
icei
nfl
uencessuppl
y
d.Obt
aina95%conf
idencei
nter
val
forpoi
ntest
imat
es
4. Supposet
hatar
esear
cherest
imat
esaconsumpt
ionsf
unct
ionandobt
ainst
he
f
oll
owi
ngr
esul
ts:
C15 0.
81Yd n19
2
(
3.1) (
18.
7) R 0.
99

wher
eC=Consumpt
ion,
Yd=di
sposabl
eincome,
andnumber
sint
hepar
ent
hesi
sar
ethe‘
t-
r
ati
os'

a. Testt
hesi
gni
fi
cantofYdst
ati
sti
cal
l
yusi
ngt
-r
ati
os
b.Det
ermi
net
heest
imat
edst
andar
ddev
iat
ionsoft
hepar
amet
erest
imat
es
5. Thefol
l
owingresul
tshav
ebeenobtai
nedfrom asimpleof11obser
vat
ionsont
he
val
uesofsal
es(Y)ofafi
rm andt
hecorr
espondingpri
ces(X)
.
X519.
18

Y217.
82

X23,
134,
i
543

XY 1,
296,
ii
836

Y2539,
i 512

a. Est
imat
ether
egr
essi
onl
i
neofsal
eonpr
iceandi
nter
prett
her
esul
ts
b. Whatisthepar
tofthevari
ationi
nsal
eswhi
chi
snotexpl
ainedbyt
her
egr
essi
on
l
ine?
c. Esti
matethepr
iceelast
ici
tyofsal
es.
6. Thefol
l
owingtabl
eincl
udest
heGNP(
X)andt
hedemandf
orf
ood(
Y)f
oracount
ryov
er
teny
earsper
iod.
y
ear 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
Y 6 7 8 10 8 9 10 9 11 10
X 50 52 55 59 57 58 62 65 68 70
a.Est
imat
ethef
oodf
unct
ion
b.Comput ethecoeffici
entofdeterminat
ionandf indt
heexplai
nedand
unexplai
nedv ar
iat
ioninthef oodexpenditure.
c. Comput ethestandarderroroftheregressioncoeff
ici
ent
sandconductt
est
ofsignif
icanceatthe5%l evelofsi
gnif
icance.
7. Asampl
eof20obser
vat
ioncor
respondi
ngt
other
egr onmodelYiXiUi
essi

gav
ethef
oll
owi
ngdat
a.
8.St
ateandexpl
aint
heassumpt
ionsunder
lyi
ngt
hemet
hodof(
ordi
nar
y)l
eastsquar
es.
9.Di
ff
erent
iat
eamongt
her
oleoft
-t
estandF-
testi
nEconomet
ri
cs
10.
Thefol
l
owingresul
tshavebeenobtainedfr
om asampleof11obser
vat
ionsont
he
val
ueofsal
es(
Y)ofafir
m andcor
respondi
ngpri
ces(
X).

X=519.
18, Y=217.
82, ΣXi2=3,
134,
543, ΣXiYi=1,
296,
836
a.Est imatethefoodf uncti
onY=β0+β1X+uandi nterpretyourresult
.
b.Comput ethepriceelasti
cit
yofsal esattheirmeanv al
ueandi nterpretyour
result
.
c.Comput eR2andi nterpr
etther esult.
d.Comput ethestandar derr
oroft her egressi
onest i
matesandconductt estsof
signif
icanceatthe5%si gnif
icantlevelandint
erpretit
.
e.Fi ndthe95%conf idenceinterval f
orthepopul ationpar ameter
sandi nt
erpreti
t
f. Predictthevalueofsal esgivenpr iceisBirr10andgi v
eeconomi cmeani ng

11.
Thef
oll
owingtabl
econtai
nsobservat
ionsonthequant
it
ydemanded(
Y)ofacer
tai
n
commodit
y,i
tspri
ce(X1)andconsumers’i
ncome(X2)
.
Y 100 75 80 70 50 65 90 100 110 60

X1 5 7 6 6 8 7 5 4 3 9

X2 1000 600 1200 500 300 400 1300 1100 1300 300

a)Est
imatethepar
amet er
sbyOLS
b)Computethecoeff
ici
entofmulti
pledeter
minati
onandint
erprett
heresul
t
c)Fi
ndthevari
anceofslopeparametersandalsothei
rst
andarderr
ors
d)Conductt
estsofsi
gnifi
canceat5%level
e)Const
ructt
he95%conf
idencei
nter
val
sfort
hepopul
ati
onpar
amet
ers
12.
Thef ol
l
owingresult
swereobtai
nedfrom asampleof12fi
rmsont hei
rout
put
(Y)
,l nput(
abori X1)andcapi
tali
nput(
X2),
measuredinarbi
tr
aryuni
ts.

Y=753 Y2=48,139 YX1=40,830

X1=643 X1 =34,
2
843 YX2=6,
796
X2=106 X22=976 X1X2=5,
779
a.Esti
mat eYonX1andX2andgi veeconomi cmeani ngofcoef f
ici
ents?
b.Giv
ent hefoll
owingsampl evaluesofout put(Y),
comput ethestandarderr
ors
oftheesti
matesandtesttheirstat
ist
icalsigni
fi
cance.
c.Fi
ndt hemul t
ipl
ecorrelati
oncoef fi
cientandt heunexpl ainedvariat
ionin
output
d.Construct99percentconf i
dencei ntervalsforthepopul ati
onparamet er
s.
e.Esti
mat etheparti
alr
egr essioncoef fi
cients,t
heirstandarderrors,andthe
2
adj
ustedandunadj ustedR v al
ues.
13.
Supposet
hatar
esear
chuseda20y
ear
sdat
aoni
mpor
tsandGDPofEt
hiopi
a.

Appl
yi
ngOLSt
otheobser
vat
ionssheobt
ainedt
hef
oll
owi
ngi
mpor
tfunct
ion.

U2=573,
069


UtUt 
1






Wher
eM =i
mpor
t,G=GDP
UsetheDurbanWatsontesttoexaminethepr
obl
em ofaut
ocorr
elat
ion.
14.
Usi
ngtheout
putbel
ow,cal
cul
atetheOLS-
esti eofβandα(
mat whi
chareshadedwi
th
bl
ack)
.

15.
Wemaybeconcer
nedthatt
hebiv
ari
atemodel
isi
nappr
opr
iat
e.Consi
dert
he
ext
endedmul
ti
var
iat
eregr
essi
on

l
n_wagei=α+βt
enur
ei+γ1Bl
ueCol
lai+γ
r 2Whi
teCol
lai+ϵ
r i
 where Bl
ueColl
arand Whi teColl
arare mutual
ly excl
usive dummy
vari
abl
esequaltooneift
heindivi
dual
’
soccupati
onisclassif
iedasblue
coll
arandwhitecoll
ar,r
especti
vel
y.Theesti
mat i
onout putfrom t
his
regr
essi
onisi
ncludedbel
ow.

A.Gi veaninterpret
ationoftheest imateont enure.
B.Cal culat
ea90per centconf i
denceintervalforβ.Giveani nter
pret
ati
onofthis
i
nterval.
C.Testt hehy pothesisH0: β=0. 05ont he5per cent-
lev
el.
16.
Wesayt hattheOLS- est
imat oris’unbi
ased’ and’ consi
stent’
.Explai
nthe
dif
ferencebet weentheset woconcept s.
17.
Explai
ninwor dshowOLSf i
ndst heregressionli
ne.
18.
TheGauss- Mar kovtheorem t el
lsusthatt heOLS- esti
mat ori so’
sal eff
ici
ent
’.
Explai
nwhati smeantbyt hisandwhatassumpt ionsarenecessar yf
orittobetr
ue.

19.Supposei
nther
egr
essi
onmodel
:yi
=β0
+βx
1i
+ε,
i
̂
β1 =0.
2, ()
̂
seβ1 =0.
1,t
a =2
2
.
306

at8dfand5%l
evel
ofsi
gif
icance

a.Canwer
ejectt
henul
lhy
pot
hesi
s:β1=0?Why
?

b.Doy
out
hingt
hatt
hev
ari
abl
exii
ssi
gni
fi
cant
?How?

20.Ev
aluat
ethest
atement“
Int
hemodel
:y=a+bx+cz+e,i
f z= x,wher
ea,bandcar
e
 parameterstobeesti
mat edandedi
sturbancet
erm sati
sfy
ingcl
assi
calassumpti
onsthen
mulit
icol
inear
it
ycoul
dn’tbeaseri
ousproblem”(
hint
:arguewhetheri
tiscorr
ectst
atement
ornot)
21.
Supposear egressionresultisgiveni nthefoll
owingtablewheredependent
vari
ableisnat urall
ogar i
sm ofhour lywageswhi l
eageandeducationareageof
respondentandy earsofschool ingappearasi ndependentvar
iabl
es.Moreover
,
dummyf orgenderv ar
iableisincludedasexpl anator
yv ar
iabl
ewhichassumes1
formaler espondentand0f orf emal e(9pts)
.
22.
Var
iabl
e 23.
coef
fi
cient 24.
Standar
d 25.
t-r
ati
o
error
26.
Cont
ant 27.
-1.
09 28.
0.38 29.
2.88
30.
mal
e 31.
0.13 32.
0.03 33.
4.47
34.
Age 35.
0.09 36.
0.02 37.
4.38
38.
educ 39.
0.18 40.
0.05 41.
3.66
2 2
R =0.691, adj .
R =0. 692,N=545, F-
criticalval
ue=2.60at3dfnominatorand
341dffordenomi nator
I
. Howdoy ouinterprett
hecoeffi
cient0. 13f orthemaledummyand0. 09for
age?
II
. Doy outhinkthatwomenar esyst emat icall
yunderpaidcompar
edtomen?
I
II
. Testt hejointhypothesi
sthatgender ,ageandeducat i
ondonotaf
fecta
Peron’
swage