Purchase Solution

Multiple Regression and Model Building

Not what you're looking for?

Ask Custom Question

Questions you need to answer:

1. What factors appear to influence beverage sales at the Westar?
2. Should Westar consider re-hiring the piano player for the Pour Nous Lounge?
3. How about happy hour food?
4. Can you suggest a new forecasting method for Jane to try out?
5. Explain what is Multicollinearity

CASE:
The Westar hotel, a 250-room business hotel located in a resort area of Colorado Rockies, has two lounges on its premises: the Pour Nous Lounge, a pub-style lounge that seats about 50, and Gaiete, a 200-seat nightclub.

Jane Bartley is a staff analyst for Westar's corporate parent, charged (among other things) with predicting and analyzing beverages sales (liquor, beer, and wine, primarily) for the chain's business hotels. In fact, her deadline for producing some sort of forecast for Westar beverage sales was the end of the week. After years in the business, Jane had seen a lot of approaches to forecasting beverages sales, but none that struck her as particularly helpful. Most of them used only room occupancy as a basis for forecasting.

Not that it was unreasonable to consider room occupancy: clearly, the number of people staying in the hotel might well influence beverage sales. Further, hotel occupancy was (at least to some extent) predictable, as people made advance reservations, so it could be used for forecasting. In the case of the Westar, though, Jane suspected other things were involved, and the existing corporate forecasting methods (such as they were) made no allowance for them. Specifically Jane wondered if season, promotional food offered during happy hour, and the presence or absence of a piano player in the Pour Nous Lounge might be factors to consider. She was able to gather data for the most recent 30 months on some relevant variables described below.

The promotional food had been discontinued after 12 month and the piano player after month 13 as the previous management attempted to cut costs. Data from month 14 and 15 were unavailable because a change in ownership about them had led to some changes in record-keeping techniques.

Jane suspected that with these data she could develop a better understanding of the factors that really influenced beverages sales at the Westar and perhaps even influence her department to consider some new and better ways to do the forecasting.

See attached excel file for the Dataset,

Purchase this Solution

Solution Preview

Multiple Regression and Model Building - Westar Beverage Sales Case Study

BEWARE OF MULTICOLLINEARITY.

The statistical analysis of the data involves Multiple Regression analysis.

Questions you need to answer:
1. What factors appear to influence beverage sales at the Westar?
First, I looked at correlations between the variables. ROOMS is not significantly correlated with any of the other variables. FOODNITE is very highly correlated with FOODTOT (because FOODPUB didn't contribute much to food sales). I would suggest only looking at FOODTOT, not the other food variables.
What variables are correlated with each other? (Table has correlation coefficients for the pairs of variables.)

BEVTOTAL BEVPUB BEVNITE FOODTOT FOODPUB FOODNITE ROOMS
BEVTOTAL 1 -0.364843176 0.971249319 0.375682013 -0.429522088 0.416438181 -0.312834383
BEVPUB 1 -0.576008264 -0.408167671 0.675388854 -0.486495263 0.134249334
BEVNITE 1 0.434199015 -0.549794259 0.490008959 -0.308983081
FOODTOT 1 -0.410014969 0.988870652 0.013952762
FOODPUB 1 -0.541148847 0.084687036
FOODNITE 1 -0.000949807
ROOMS 1

Significant correlations (r > 0.374) are in italics.

"Rooms" is not statistically significantly correlated with any of the other variables.
Then I looked at how the variables changed over time. Total beverage sales (and BEVNITE) tended to peak in spring (March-April) and in December. There was a decline after the food/piano was taken away, but sales recovered. BEVPUB didn't show quite as much seasonality (no clear, high spikes in sales), and there was a slight decline after the food/piano was taken away, but these sales did not contribute much to the total beverage sales. FOODTOT didn't really have any seasonality either, and increased a little after the end of food/piano. ROOMS had peaks in March and the summer months.

BEV TOTAL peaks at months 4, 12, 16, 24, 27 (April, December, April, December, March) sharp decline when promotional food was taken away, but then there was a recovery (and the decline could have been ...

Purchase this Solution


Free BrainMass Quizzes
Know Your Statistical Concepts

Each question is a choice-summary multiple choice question that presents you with a statistical concept and then 4 numbered statements. You must decide which (if any) of the numbered statements is/are true as they relate to the statistical concept.

Measures of Central Tendency

This quiz evaluates the students understanding of the measures of central tendency seen in statistics. This quiz is specifically designed to incorporate the measures of central tendency as they relate to psychological research.

Terms and Definitions for Statistics

This quiz covers basic terms and definitions of statistics.

Measures of Central Tendency

Tests knowledge of the three main measures of central tendency, including some simple calculation questions.