Construct a line chart for your given brand of beer. Comment on any features of interest.
Estimate a model of volume of sales using a trend component. In other words, estimate the
Where yt = Volume of sales in period t
T = Trend
Comment on your findings. I.e. what is the interpretation of the coefficient estimates? How do your
findings for the trend variable compare to your observations in step 1? Are the variables statistically
significant? What about the overall model fit? [5 marks]
Note: To complete this step, you must create a trend variable.
Refine the model you created in step 2 by adding a seasonal index. In other words, estimate the
= + +
Where SI = Seasonal Index
Are there any months that exhibit above average sales? What do you think is the reason? How do
your findings compare to your observations in step 1? You may wish to report the seasonal indices
you constructed. Comment on your findings. [10 marks]
Note 1: To complete this step, you must create a seasonal index.
Note 2: Including the seasonal index may cause the intercept to become statistically insignificant.
However, the intercept should be retained as the line chart indicates that the data does not pass
through the origin.
In the year 2005 and later in 2012, the outbreak of a “killer yeast strain” – Saccharomyces Cerevisiae
- affecting the beer brewing process had a devastating impact. Known as ‘Gastroenteritis’ , the fear
of the virus quickly spread throughout the world causing people to avoid consumption of beer and
yeast-related beer products. Many (not all) brands of beer use yeast in their brewing process. How
do you think this yeast strain would affect the sales of your assigned brand? At the same time,
certain brands of beer do not use yeast in their production. How do you think their sales will be
Refine the model you created in step 3 by adding a dummy variable to represent the occurrence of
‘Gastroenteritis’. In other words, estimate the following:
Having accounted for various time series effects, your client is now interested in how pricing affects
sales. You must investigate own price effects as well as the price effects of various competitors. Use
either forward or backward selection to determine which competitors are relevant. Estimate the
= + +
+ â‹¯ +
Where Pt = Own price in period t
= Price of the kth competitor in period t
What is the effect of increasing the price of your own brand? What about the effect from
competitors? Which competitors are relevant and which ones are not? Why do you think your brand
is affected by some competitors and not others? [15 marks]
Hint: Different brands usually compete in different categories.
Your client believes that sales may be affected by other savoury items and not just other brands of
beer. In particular, your client believes that peanuts – a ‘must-have’ accompaniment in beer
drinking, may affect sales as they may be regarded as a complementary good. Investigate the
impact of peanuts on sales of your brand. Estimate the following:
= + +
+ â‹¯ +
IMPORTANT: at this stage of the regression you will need to refine your regression to only include
explanatory variables you deem to be important. Excel (unfortunately) will not allow you to include
more than 16 variables in the regression.
Final Model: Now that you have done your analysis, recommend a model for your brand
performance and justify your recommendation. Remember foremost your client is not an
Econometrician! [5 marks]
Robustness Section: ensure your final model is robust to the assumptions underpinning the theory
of Financial Modelling. Include only plots and tests you deem important and essential. You may also
want to discuss the limitations of your regression and any issues related to this final model (i.e. are
there other variables worth considering? Is there collinearity in your regression? etc.)
IC = Average price of peanuts in period t
Is your client’s theory valid? Are beer and peanuts complementary goods? [5 marks]
Your client wants you to investigate the merits of different shelfing positions and their effects on
sales. Currently liquor stores are charging different rents depending on the shelf position chosen.
Three(3) positions are available : lower shelf (cheaper pricing), middle shelf (premium pricing) and
upper half shelf (medium pricing). To test the effects on different shelfing positions, your client has
been changing shelf positions throughout the sample period. Your client now wants you to
investigate empirically the merits of shelf positions, and conclude whether in fact there is a
relationship between shelf position and sales. Construct the appropriate number of dummy
variables and comment on your results. [15 marks]
= + +
= Disease dummy variable for the 2003 period
= Disease dummy variable for the 2010 period
What is the overall impact of the disease between the two periods? How do your findings compare
to your observations in step 1? Comment on your findings. [10 marks]
Note: To complete this step, you must create dummy variables.