Applied Econometrics For Managers: Problem Set #1: Section - B Group Number - 2

Applied Econometrics for Managers: Problem Set #1
SECTION – B
GROUP NUMBER – 2
GROUP MEMBERS:
Sajal Kumar (MBA19223)
Vivek Kedia (MBA19244)
Vijay Dhayanandh M (MBA19158)
Kumar Safal (MBA19197)
Saharsh Maheshwari (MBA19138)
Dataset used: “macchiavello_morjaria_aer_2015.dta
Variables used in the dataset:
 weekq (weekly quantity of roses sold, average, in thousands)
 price (average price per stem)
 age (days since first transaction between a buyer and a seller)
Analysis:
(a) Bring the dataset into R.
(b) Rename the “age” variable as “age_rel”.

(c) Provide the summary statistics (mean, std dev, median, min, max, and number of
Observations n) for the variables weekq, price, and age.
(d) Estimate the following relationships using ordinary least squares method and provide the results
(coefficients table along with the Adjusted R-squared)
1) weekq = β0 + β1price + e (level-level regression)

Interpretation:
 In this model, 1 Euro increase in price of roses per
stem will decrease the quantity sold by 3384.5
(~3385) roses.
 We took 95% confidence interval in the given
example. For 95% confidence interval, Pr value
becomes equal to 0.05. As Pr value is less than 0.05,
we can say that the data is statistically significant. We
can reject the null hypothesis. This means β1 is not
equal to 0.
 Adjusted R-square value is 0.02248, we can say that
through this regression model 2.248% of the variation
in the data can be explained.
2) log(weekq) = β0 + β1log(price) +e (log-log regression)
Interpretation:
 In this model, 1% increase in price of roses will decrease the
quality sold in a week by 0.96625%
 We took 95% confidence interval in the given example. For 95%
confidence interval, Pr value becomes equal to 0.05. As Pr value
is less than 0.05, we can say that the data is statistically
significant. We can reject the null hypothesis. This means β1 is
not equal to 0.
 Adjusted R-square value is 0.1297, we can say that with this
model 12.97% of the variation in the data can be explained.
3) log(weekq) = β0 + β1price + e (log-level regression)
Interpretation:
 In this model, 1% increase in price of the roses per stem will decrease the quantity sold in a week by 419 roses
 We took 95% confidence interval in the given example. For 95% confidence interval, Pr value becomes equal to
0.05. As Pr value is less than 0.05, we can say that the data is statistically significant. We can reject the null
hypothesis. This means β1 is not equal to 0.
 Adjusted R-square value is 0.01865, we can say that with this model 1.865% of the variation in the data can be
explained.
4) weekq = β0 + β1log(price) + e (level-log regression)
Interpretation:
 In this model, 1 Euro increase in price of the roses per stem

will decrease the quantity sold in a week by 7.515%
 We took 95% confidence interval in the given example. For
95% confidence interval, Pr value becomes equal to 0.05. As
Pr value is less than 0.05, we can say that the data is
statistically significant. We can reject the null hypothesis. This
means β1 is not equal to 0.
 Adjusted R-square value is 0.1445, we can say that with this
model 14.45% of the variation in the data can be explained
(e) Create a new variable: pricesq =

price2
(f) Estimate the relationship, and provide the results:
weekq = β0+β1price+β2pricesq+β3age−rel +e
Interpretation:
 In this model, 1 Euro increase in price of the roses per
stem will decrease the quantity sold in a week by
7594.77 roses on average
 1 day increase in transaction between a buyer and a
seller will increase the quantity sold per week by 58
roses on average
 1-unit price_square increase will increase the quantity
sold in a week by 88 roses on average
 We took 95% confidence interval in the given example.
For 95% confidence interval, Pr value becomes equal to
0.05. As Pr value is less than 0.05, we can say that the
data is statistically significant. We can reject the null
hypothesis. This means β1 is not equal to 0.
 Adjusted R-square value is 0.04127, we can say that
with this model 4.127% of the variation in the data can
be explained.
Final intuition about regression results
 It can be inferred from the analysis that quantity sold is decreasing as price per stem is increasing.
 In this data set, Log-Level Regression model explains 14.45 variation of the data which is highest among all
four models. So, we can say that Log- Level model is most suitable for the given dataset.
 It can also be inferred that quantity sold is decreasing exponentially as price of roses increasing.

Applied Econometrics For Managers: Problem Set #1: Section - B Group Number - 2

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Applied Econometrics For Managers: Problem Set #1: Section - B Group Number - 2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Econometrics For Managers: Problem Set #1: Section - B Group Number - 2

Uploaded by

Copyright:

Available Formats

Applied Econometrics for Managers: Problem Set #1

(a) Bring the dataset into R.

(b) Rename the “age” variable as “age_rel”.

1) weekq = β0 + β1price + e (level-level regression)

2) log(weekq) = β0 + β1log(price) +e (log-log regression)

3) log(weekq) = β0 + β1price + e (log-level regression)

4) weekq = β0 + β1log(price) + e (level-log regression)

 In this model, 1 Euro increase in price of the roses per stem

(e) Create a new variable: pricesq =

Final intuition about regression results

You might also like