Professional Documents
Culture Documents
Multiple Regression Analysis: 2. Statistical Inference (Hypothesis Testing)
Multiple Regression Analysis: 2. Statistical Inference (Hypothesis Testing)
2. Statistical Inference
(Hypothesis Testing)
Var ˆ j
SSTj 1 R j2
se ˆ j
is the (estimated)
standard error of ˆ j
ˆ j
~ Normal 0,1
j
sd ˆ j
ˆ j is distributed normallybecause it
is a linear combination of the errors
2011-3-28 Lectured by Dr. Jin Hongfei 10
The t Test
Under the CLM assumptions
ˆ j j ~t
se ˆ j
n k 1
Fail to reject
reject
0 c
2011-3-28 Lectured by Dr. Jin Hongfei 19
One-sided (tailed) vs Two-sided
Because the t distribution is symmetric, testing H1:
j < 0 is straightforward. The critical value is just
the negative of before
We can reject the null if the t statistic < –c, and if
the t statistic >–c then we fail to reject the null
For a two-sided test, we set the critical value
based on /2 and reject H1: j 0 if the absolute
value of the t statistic > c (|t| > c)
H0: j = 0 H1: j ≠ 0
fail to reject
reject reject
-c 0 c
2011-3-28 Lectured by Dr. Jin Hongfei 21
Summary for H0: j = 0
Unless otherwise stated, the alternative is
assumed to be two-sided
If we reject the null, we typically say “xj is
statistically significant at the level”
If we fail to reject the null, we typically say
“xj is statistically insignificant at the
level”
t
ˆ j 0 j , where
se ˆ j
0 0 for the standard test
j
F
SSRr SSRur q
, where
SSRur n k 1
r is restricted and ur is unrestricted
2011-3-28 Lectured by Dr. Jin Hongfei 36
The F statistic
The F statistic is always positive, since the
SSR from the restricted model can’t be less
than the SSR from the unrestricted
Essentially the F statistic is measuring the
relative increase in SSR when moving from
the unrestricted to restricted model
q = number of restrictions, or dfr – dfur
n – k – 1 = dfur
reject
0 c F
2011-3-28 Lectured by Dr. Jin Hongfei 39
Example (4.9)
bwght = β0 + β1cigs + β2parity +
β3faminc + β4motheduc + β5fatheduc + u
bwght – birth weight (lbs), cigs – avge. no.
of cigarettes smoked per day during
pregnancy, parity – birth order, faminc –
annual family income, motheduc – years of
schooling of mother, fatheduc – years of
schooling of father.
2011-3-28 Lectured by Dr. Jin Hongfei 40
Example (4.9) continued
Test whether, controlling for other factors,
parents’ education has any impact on birth
weight
H0: β4 = β5 = 0 H1: H0 not true
q = 2 restrictions
Restricted model:
bwght = β0 + β1cigs + β2parity +
β3faminc + u
SSRur
2011-3-28 Lectured by Dr. Jin Hongfei 42
Example (4.9) continued
SSRr
2011-3-28 Lectured by Dr. Jin Hongfei 43
Example (4.9) continued
Reject if F > F2,1185 = 3.00
( SSRr SSRur ) / q
F
SSRur /(n k 1)
(465166.792-464041.135)/2
1.44
464041.135 /(1191-5-1)
so we do not reject H0. Parental education
is not a significant determinant of birth
weight.
2011-3-28 Lectured by Dr. Jin Hongfei 44
Example (4.9) easier way!
Use the test command in Eviews
Command here is: eq1.wald c(5)=0, c(6)=0
Type the following commands in command
area:
Equation eq1.ls bwght c cigs parity faminc motheduc fatheduc
eq1.wald c(5)=0, c(6)=0
F
R q R
2 2
, where again
ur r
1 R n k 1
2
ur
ˆ1 ˆ 2
t
se ˆ1 ˆ 2
2011-3-28 Lectured by Dr. Jin Hongfei 53
Testing Linear Combo (cont)
Since
se ˆ1 ˆ 2 Var ˆ1 ˆ 2 , then
Var ˆ ˆ Var ˆ Var ˆ 2 Cov ˆ , ˆ
1 2 1 2 1 2