Questions and Answers of Essentials Of Marketing Analytics

How would you incorporate nonlinearities and/or interactions in a logistic regression model?
The file Healthcaredata.xlsx gives the age, party affiliation, and income of 300 people. You are also told whether they favor Obamacare. Develop a model to predict the chance that a person favors
The file Logitsubscribedata.xls gives the number of people in each age group who subscribe and do not subscribe to a magazine. How does age influence the chance of subscribing to the magazine?
For 1,022 NFL field goal attempts the file FGdata.xlsx contains the distance of the field goal and whether the field goal was good. How does the length of the field goal attempt affect the chance to
Manipulation of Equation 1 shows the probability that a dependent variable equals one and is estimated by the following equation: eBo+Blx1+-Baxn 1+eo+B1x1+Baxn
You can use programs such as SAS, SPSS, and StatTools to easily estimate the logistic regression model and test the significance of the coefficients.
The coefficients in Equation 1 are estimated using the method of Maximum Likelihood.
To predict the probability p that a binary-dependent variable equals 1, use the logit transformation: (1) Ln 1-p Bo+Bx,+Bx+... Box
When the dependent variable is binary, the assumptions of multiple linear regression are violated.
How Palisade's StatTools program can easily be used to estimate a logistic regression model and test hypotheses about the individual coefficientsHow to interpret the coefficients in a logistic
How to interpret the coefficients in a logistic regression model
How the Excel Solver may be used to implement the technique of maximum likelihood to estimate a logistic regression model
Why multiple linear regression is not equal to the task of predicting a binary dependent variable
The soda.xlsx fi le (see Figure 16-13) gives a consumer’s ranking on an orthogonal design with 12 product profi les involving a comparison of Coke and Pepsi. The attributes and levels are as
Determine how various attributes impact the purchase of a car. There are four attributes, each with three levels: Brand: Ford = 0, Chrysler = 1, GM = 2 MPG: 15 MPG = 0, 20 MPG 1, 25 MPG = 2
Choice-based conjoint analysis ascertains the importance of attributes and the ranking of levels within each attribute by asking the consumer to simply choose the best of several product profiles
Adaptive/hybrid conjoint analysis is often used when there are many product attributes. After having the consumer answer questions involving ranking attributes and levels, the software adapts to the
A conjoint simulator can predict how changes in existing products (or intro- duction of new products) will change product market shares.
Multiple linear regression (often using dummy variables) can easily be used to rank the importance of attributes and the ranking of levels within each attribute.
To make the estimates of each attribute's importance and level preferences more accurate, it is usually preferable to make the correlation between any pair of attributes in the product profiles close
In full profile conjoint analysis, the consumer is asked to rank a variety of product profiles.
Conjoint analysis is used to determine the importance of various product attributes and which levels of the attributes are preferred by the customer.
The file Quarterly.xlsx discussed in Chapter 14 gives quarterly revenues and quarterly profits for several U.S. companies. Use a neural network to predict next quarter's revenues based on the last
For the neural network you found in Problem 3, predict total revenue for the following two movies:a. Movie 1: Week 1 revenue $50 million; Week 2 revenue $50 millionb. Movie 2: Week 1 revenue: $80
The file movierevenue.xls gives revenue for several movies during weeks 1 and 2 of release as well as their total revenue. Use a neural network to predict total revenue from week 1 and week 2
The file windchill.xlsx gives the wind chill index (a measure of the danger from wind and temperature) for a variety of winds and temperatures. Use a neural network to predict the wind chill index
What independent variables and dependent variables would be used to train a neural network that can drive a car? Hint: Assume the car has video cameras that can see in front of, in back of, and to
If the MAD for a neural net is not much better than the MAD for a multiple linear regression, then you can be confident that the relationship between the dependent and independent variables can be
If the MAD for the testing set is far larger than the MAD for the training set, then the neural net is overfitting the data and should not be used for forecasting.
The Palisades add-in, NeuralTools, makes it easy to fit a neural set to data.
Neural nets have been used to forecast stock prices, sales of consumer pack- aged goods, and even to drive a car!
You can use neural networks to search for a nonlinear relationship that best explains the relationship between a dependent variable and a set of indepen- dent variables.
Suppose at the end of 2007 you were predicting housing starts in Los Angeles for the years 2008 and 2009. Why do you think Winter's Method would pro- vide better forecasts than multiple regression?
Use Winter's Method to forecast one-quarter-ahead revenues for Amazon. com.
Use Winter's Method to forecast one-quarter-ahead revenues for Apple.
Use Winter's Method to forecast one-quarter-ahead revenues for Home Depot.
Use Winter's Method to forecast one-quarter-ahead revenues for Coca-Cola.
Use Winter's Method to forecast one-quarter-ahead revenues for Wal-Mart.
Forecasts for k periods ahead at the end of period t are made with Winter's Method using Equation 4: (4) F-L(T)
Winter's Method is an exponential smoothing method that updates the base, trend, and seasonal indices after each equation: (1) L-alp(x)/(s)+(1-alp) (L * T) (2) T-bet(L/L) + (1-bet)T (3) S, =
Exponential smoothing methods update time series parameters by computing a weighted average of the estimate of the parameter from the current observa- tion with the prior estimate of the parameter.
Based on the data in the file airlineniles.xlsx from Chapter 12, use the Ratio to Moving Average Method to forecast airline miles for the remaining months in 2012.
The file Walmartdata.xls contains quarterly revenues of Wal-Mart during the years 1994-2009. Use the Ratio to Moving Average Method to forecast revenues for Quarters 3 and 4 in 2009 and Quarters 1
You can apply the Ratio to Moving Average Method to monthly data as well by following the same process but use 12-month moving averages to eliminate seasonality.
Applying the Ratio to Moving Average Method involves the following tasks: Compute four-quarter moving averages and then determine the cen- tered moving averages. Fit a trend line to the centered
Determine the trend and seasonality for the years 1995-2004.
Determine the trend and seasonality for the years 1981-1990.
Determine the trend and seasonality for the years 1970-1980.
You can often use the following equation to predict sales of a product: Predicted period 1 Sales = Base (Trend) (Seasonal Index for Month () The following exercises use the file airlinedata.xlsx,
You can often use seasonality and trend to predict sales by using the follow- ing equation: Predicted Period t Sales - Base +Trend Period Number + Seasonal Index for Month t
Using a 12-month or 4-quarter moving average chart enables you to easily see the trend in a product's sales.
The file Promotiondata.xlsx contains monthly sales (in pounds) of ice cream at a Kroger's supermarket for three years. The file also tells you when promo- tions occurred. Promotions are known to
The file Dinner.xls contains a model to predict daily dinner sales at a well- known chain restaurant. Column Q of the worksheet outliers removed contains the final forecast equation.a. Explain in
If the signs of residuals change much less than half the time, then forecast errors are not random.
You can use outliers to spot omitted special factors.
Using the Excel Solver you can often forecast daily demand with the model Base Level + Day of Week Effect + Month Effect + Effect of Special Factors.
The Excel Solver can be used to mimic regression analysis and work around the 15 independent variable limitations of Excel's Regression tool.
Does our regression model for predicting auto sales exhibit autocorrelation, non-normality of errors, or heteroscedasticity?
The file USmacrodata.xlsx contains US. quarterly GNP, Inflation rates, and Unemployment rates. Use this file to perform the following exercises:a. Develop a regression to predict quarterly GNP growth
The file Oreos.xlsx gives daily sales of Oreos at a supermarket and whether Oreas were placed 7 from the floor, 6" from the floor, or 5" from the floor. How does shelf position influence Oreo sales?
Determine for your regression predicting computer sales whether the residuals exhibit non-normality or heteroscedasticity.
The file Cardata.xlsx provides the following information for 392 different car models: Cylinders Displacement Horsepower Weight Acceleration Miles per gallon (MPG) Determine an equation that can
The file Basebal196.xlsx gives runs scored, singles, doubles, triples, home runs, and bases stolen for each major league baseball team during the 1996 season. Use this data to determine the effects
The file Countryregion.xlsx contains the following data for several under- developed countries: Infant mortality rate Adult literacy rate Percentage of students finishing primary school Per capita
For 12 straight weeks, you have observed the sales (in number of cases) of canned tomatoes at Mr. D's Supermarket. (See the file Grocery.xlsx.) Each week, you keep track of the following:a. Was a
Fizzy Drugs wants to optimize the yield from an important chemical process. The company thinks that the number of pounds produced each time the process runs depends on the size of the container used,
If independent variables are highly correlated, then their coefficients in a regression may be misleading. This is known as multicollinearity.
You can check for the presence of autocorrelation in a regression based on time series data by examining the number of sign changes in the residuals; too few sign changes indicate positive
By adding the product of two independent variables (say X, and X,) as a new independent variable, you can test whether X, and X, interact in their effect on Y.
Qualitative independent variables are modeled using indicator variables. By adding the square of an independent variable as a new independent vari- able, you can test whether the independent variable
Approximately 68 percent of predictions from a regression should be accurate within one standard error and approximately 95 percent of predictions from a regression should be accurate within two
Independent variables with p-value greater than .05 should be deleted, and the regression should be rerun until all independent variables have p-values of .05 or less.
The Coefficients portion of the output gives the least squares estimates of BBB A Significance F in the ANOVA section of the output less than .05 causes you to reject the hypothesis of no linear
You can run a regression analysis using the Data Analysis Tool.
Violation of these assumptions can invalidate the p-values in the Excel output.
The error term is required to satisfy the following assumptions: The error term is normally distributed. The variability or spread of the error term is assumed not to depend on the value of the
The multiple linear regression model models a dependent variable Y as B.+ BX, BX, ...BX+ error term.
NFL teams play 16 games during the regular season. Suppose the standard deviation of the number of games won by all teams is 2, and the correlation between the number of games a team wins in two
The file Printers.xlsx contains daily sales volume (in dollars) of laser print- ers, printer cartridges, and school supplies. Find and interpret the correlations between these quantities.
The file Parking.xlsx contains the number of cars parked each day both in the outdoor lot and in the parking garage near the Indiana University Kelley School of Business. Find and interpret the
Find the trend line to predict soda sales from daily bowl sales.
The file GNP.xls (available on the companion website) contains quarterly GNP data for the United States in the years 1970-2012. Try to predict next quarter's GNP from last quarter's GNP. What is the
The file Electiondata.xlsx (available on the companion website) contains, for several elections, the percentage of votes Republicans gained from voting machines (counted on election day) and the
You know that 32 degrees Fahrenheit is equivalent to 0 degrees Celsius, and that 212 degrees Fahrenheit is equivalent to 100 degrees Celsius. Use the trend curve to determine the relationship between
The file Housedata.xlsx (available on the companion website) gives the square footage and sales prices for several houses in Bellevue, Washington. Use this file to answer the following questions:a.
The file Dell data.xlsx (available on the companion website) contains monthly returns for the Standard & Poor's stock index and for Dell stock. The beta of a stock is defined as the slope of the
If x is k standard deviations above the mean, you can predict y to be rk stan- dard deviations above the mean.
Correlation may be computed with the Analysis ToolPak or the CORREL function.
Given two variables x and y, the correlation r (always between-1 and +1) between x and y is a measure of the strength of the linear association between x and y.
Approximately 95 percent of the forecasts from a least-squares line are accu- rate within two standard errors of the regression.
The R value is the fraction of variation in the dependent variable explained by variation in the independent variable.
The Excel Trendline can be used to find the line that best fits data.
Often the valuation customers have for products drops over time. This requires that retailers lower or mark down their prices over time.
Marginal analysis is helpful to solve booking limits and overbooking problems.
To adjust for the possibility of no shows, organizations need to sell more capacity than is available.
To handle the fact that organizations do not know how many high-value customers will demand a product, organizations often set booking limits to constrain low-price sales so that more capacity is
Revenue management has a much greater chance of succeeding if "fences" (such as staying over Saturday night for airlines) exist to separate high-valu- ation customers from low-valuation customers.
Revenue management enables organizations including airlines, hotels, rental car agencies, restaurants, and sports teams to increase profits by reducing the unused amount of perishable inventory
The pre-Christmas demand for Christmas cards at a local Hallmark stores is given by q=2000-300p. The demand for Christmas cards after Christmas is given by q=1000-400p. If the store pays $1 per card,
All tickets are nonrefundable. How many reservations should FBN accept on this flight?

Showing 200 - 300 of 1275