Sie sind auf Seite 1von 4

Tutorial 6:

Q2.
a) State the estimated equation and interpret the coefficients. Are the signs of the estimates
consistent with the expectations?








= =

=
= =

=
|
|

Therefore,
=
=
=
+ + =

Interpretation:
b1: For every additional year in the clocks age, auction price of clocks will increase by
$9.0014 holding all other variables constant.
b2: For every additional bidder present at the auction, auction price of clocks will increase by
$89.2904 holding all other variables constant.
b0: When a new clock is up for auction with no bidders present, the auction price of the cock
will be -$800.6894.
This requires the use of a formula we learnt back in ECON1310:

Where:

Note: When asked to calculate a confidence interval, t stat should NOT BE USED apart from in
calculating s
bi
. Remember that the t
crit
value is used in confidence intervals (NOT t
calc
) which we
need to find from the t tables.

Expectations:






Independent variable Expected relationship Model-depicted
relationship with Y
Consistent?
Age > | > | Yes
Number of bidders. > | > | Yes

b) Test if there is a significant overall relationship at the 5% LOS


Step 1: State Hypothesis
= = | | (no significant relationship b/w y all xs)
= |
Step 2: Decision Rule (always upper-tail F-test)
Reject H
0
if F
calc
> F
,k,n-k-1
= F
0.05 (2,32-2-1)
= 3.33
Step 3: Calculate Test Statistic



F
calc
= MSC/MSE = 1787236.697/42252.27953 = 42.2991781

Step 4: Decision: Reject H
0
as F
calc
> F
crit

Step 5: Conclusion: There is sufficient evidence at the 5% level of significance to conclude
that there is a significant relationship between auction price and at least one of clock age
and number of bidders.



We need to think and use some intuition here.
Age: as the antique clock gets older, (increases in age), it is likely to become more
valuable as it will most likely be rarer, thus we would expect .

Number of bidders: As the number of bidders increases, we would expect that the price
will increase as more people bidding for a scarce resource drives up its price. Thus, would
expect .

Remember, a test of overall significance uses an F-test to see if all our slope coefficients are
significant.
c) Test the hypothesis that the auction price of the clock increases as the number of bidders
increase. Use 5% LOS.

Step 1: State Hypothesis
s | (no significant relationship b/w y and X
2
)
> |
Step 2: Decision Rule (always upper-tail F-test)
Reject H
0
if t
calc
> t
,n-k-1
= t
0.05,(32-2-1)
= 1.699
Step 3: Calculate Test Statistic



t
calc
= 6.532

Step 4: Decision: Reject H
0
as t
calc
> t
crit

Step 5: Conclusion: There is sufficient evidence at the 5% level of significance to conclude
that there is a significant relationship between auction price and number of bidders.

d) Test the hypothesis that the auction price of the clock increases as the number of bidders
increase. Use 5% LOS.

Step 1: State Hypothesis
s | (no significant relationship b/w y and X
1
)
> |
Step 2: Decision Rule (always upper-tail F-test)
Reject H
0
if t
calc
> t
,n-k-1
= t
0.05,(32-2-1)
= 1.699
Step 3: Calculate Test Statistic



t
calc
= 8.177

Step 4: Decision: Reject H
0
as t
calc
> t
crit

Step 5: Conclusion: There is sufficient evidence at the 5% level of significance to conclude
that there is a significant relationship between auction price and the clock age.




e) Calculate the coefficient of multiple determination and interpret its value.
Remember the coefficient of multiple determination is the same as R
2
.
= =

= =
Thus we can calculate R
2
in two ways:
= =

Ill use the second way, this time (doesnt matter which one we use):
= =
74.47% of variation in auction price is explained by the variation in the explanatory variables
(clock age and number of bidders).

Q3. Is identical to what we did in question 1 in the tutorial.

Das könnte Ihnen auch gefallen