| Course | GB540 Economics for Global Decision Makers |
|---|---|
| Unit | Unit 3 |
| Paper type | Demand estimation exercise |
| Length | About 620 words, 5 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | MBA |
| Updated | October 2026 |
Free sample paper for GB540 Unit 3
Forty-Eight Dealer-Seasons and an Elasticity Near Minus 2.31: A Unit 3 Demand Estimation Exercise
[Student Name]
Purdue University Global
GB540: Economics for Global Decision Makers
Unit 3 Demand Estimation Exercise
[Instructor Name]
[Date]
Ruddiman Marine, its dealers and every statistic are composite and illustrative; the economics sources cited are real.
The Data
The sample covers all sixteen of Ruddiman Marine's US dealers over three seasons, 48 observations in all. Each observation is one dealer in one season.
Quantity. Motors sold by the dealer in the season. The sample mean is 138.7.
Own price. The dealer's average selling price for Ruddiman's 6-kilowatt motor, in dollars. Mean $4,300. Prices varied across dealers and seasons because dealers set their own discounts within a band.
Rival price. The price of a comparable gasoline outboard in the dealer's market, in dollars. Mean $3,210.
Income. County median household income, in thousands of dollars. Mean 86.1.
Charging slips. Marina slips with charging equipment within 40 miles of the dealer. Mean 31.4.
The Regression
Quantity was regressed on the four variables by ordinary least squares.
Own price. Coefficient minus 0.0745, standard error 0.0069, t statistic minus 10.78. Each $100 increase in the dealer's price is associated with about 7.5 fewer motors a season.
Rival price. Coefficient 0.0471, standard error 0.0118, t statistic 3.99. A $100 increase in the gasoline motor's price is associated with about 4.7 more electric motors sold.
Income. Coefficient 0.7412, standard error 0.2205, t statistic 3.36.
Charging slips. Coefficient 0.8834, standard error 0.3410, t statistic 2.59.
Fit. R-squared 0.939, adjusted R-squared 0.933, F statistic 165.5 on 4 and 43 degrees of freedom. All four coefficients have the expected sign and are significant at the 5 percent level.
The Elasticities
Each elasticity is the coefficient times the variable's mean divided by mean quantity.
Own price. Minus 0.0745 times 4,300 divided by 138.7 gives about minus 2.31. Demand is elastic, and the size is ordinary for a branded product: Tellis (1988) put the average across earlier sales studies near minus 1.76.
Rival price. 0.0471 times 3,210 divided by 138.7 gives about 1.09. The gasoline motor is a substitute, consistent with the Norwegian cross-price estimate in Unit 2, though smaller.
Income. 0.7412 times 86.1 divided by 138.7 gives about 0.46. The motor is a normal good, but sales respond less than proportionally to income.
Charging slips. 0.8834 times 31.4 divided by 138.7 gives about 0.20.
A demand equation earns trust when its elasticities agree with the separate evidence the firm already has, and minus 2.31 sits close to the Australian estimate of minus 1.91.
A Caution on Price
Dealers set their own discounts, and a dealer who expects a weak season may cut price because demand is low. When price responds to demand that way, ordinary least squares can understate how strongly buyers respond. Berry (1994) describes this problem in differentiated product markets and the use of instruments, such as cost shifters, to address it. Ruddiman has no good instrument in this sample, so minus 2.31 should be read as an estimate that could be somewhat too small in magnitude. Working (1927) made the general point that price and quantity data trace demand only when the variation in price comes from somewhere other than demand.
The Forecast
For next season, with the rival's price, county income and charging slips set at their projected values, the equation predicts about 137 motors per dealer-season at the current price of $4,400. At $4,600, it predicts 137 minus 0.0745 times 200, or about 122. Across sixteen dealers, that is about 2,192 motors at $4,400 and about 1,954 at $4,600.
Whether the higher price pays depends on cost, which the next unit estimates.
References
Berry, S. T. (1994). Estimating discrete-choice models of product differentiation. The RAND Journal of Economics, 25(2), 242-262. https://doi.org/10.2307/2555829
Tellis, G. J. (1988). The price elasticity of selective demand: A meta-analysis of econometric models of sales. Journal of Marketing Research, 25(4), 331-341. https://doi.org/10.1177/002224378802500401
Working, E. J. (1927). What do statistical "demand curves" show? The Quarterly Journal of Economics, 41(2), 212-235. https://doi.org/10.2307/1883501
What the GB540 Unit 3 assignment asks for
Unit 3 in GB540 asks you to estimate a demand equation and interpret it. Describe your data: the unit of observation, the number of observations and each variable with its units and mean. Run a regression of quantity on own price and at least two other demand factors. Report every coefficient with its standard error and t statistic, along with R-squared, adjusted R-squared and the F statistic. Convert each coefficient into an elasticity at the sample means, showing the calculation, and interpret each. Compare your estimates with other evidence you have. Discuss any reason the price coefficient might be biased and in which direction. Then use the equation to forecast quantity at the current price and at an alternative price. About five pages, covering the data, the regression, the elasticities and the forecast, is common for this exercise.
Inside the GB540 Unit 3 example
The sample defines quantity, own price, rival price, county income and charging slips, each with units and mean, with a margin note on why the means matter. The regression section reports each coefficient, such as minus 0.0745 on own price with a t of minus 10.78, and the fit statistics. Elasticities at the means follow, with the own-price figure compared with Tellis's average. The highlighted sentence notes agreement with the Australian estimate. A caution explains that dealers who expect weak demand may cut price, which can bias the estimate toward zero, citing Berry and Working, and a margin note gives the direction. The forecast shows about 137 motors per dealer at $4,400 and 122 at $4,600.
GB540 Unit 3 rubric: what earns full points
GB540 instructors grade the demand exercise on correct estimation, clear reporting and careful interpretation. The data description should define every variable with units. The regression table should report coefficients, standard errors, t statistics and fit measures. Converting coefficients to elasticities at the means, with the calculation shown, is expected. Each elasticity should be interpreted by sign and size. Comparing estimates with other evidence shows judgment. Discussing possible bias in the price coefficient, and its direction, earns credit at MBA level. A forecast at two prices shows the equation can support a decision. A high R-squared should not be presented as proof that coefficients are unbiased, and methods sources should be cited.
GB540 Unit 3 help from the desk
Many GB540 demand exercises report coefficients without units, so elasticities cannot be checked. Others present a high R-squared as proof the model is right. Students often skip converting coefficients into elasticities or compute them at arbitrary values. Possible bias from prices that respond to demand is rarely mentioned. Forecasts use only the current price, leaving the decision untested. Define each variable, report full statistics, compute elasticities at the means, interpret each, compare with other evidence, discuss the direction of any bias and forecast at more than one price. Send your dataset, and our team can help run and report the regression.
Get a GB540 Unit 3 example written to your instructions
If the Unit 3 exercise hands over a dataset, send it along with the instructions and rubric; if it asks for one, say so. The sample returns in 24-48h with the equation written out, every coefficient interpreted in its units, elasticities at the means and a forecast with its assumptions stated. No fee applies to the first one. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More GB540 papers and MBA samples
- GB540 Unit 1: Discussion Board Post
- GB540 Unit 2: Elasticity Analysis
- GB540 Unit 4: Production Cost Analysis
- GB540 Unit 5: Market Structure Comparison
- GB540 Unit 6: Pricing Strategy Paper
- GB540 Unit 7: Macroeconomic Indicator Analysis
- GB540 Unit 8: Monetary and Fiscal Policy Memo
- GB540 Unit 9: Exchange Rate Exposure Paper
- GB540 Unit 10: Firm Economic Analysis
- GB541 Unit 3: Risk Register
- GB534 Unit 3: Motivation Analysis
- GB542 Unit 3: Talent Segmentation Memo
- GB533 Unit 3: Workload Sizing Model
GB540 Unit 3 questions, answered
How does a regression coefficient become an elasticity?
For a linear demand equation, multiply the coefficient by the ratio of the variable's value to quantity, usually both taken at their sample means. A price coefficient of minus 0.0745, a mean price of about 4,356 and mean sales of 140.4 give roughly minus 2.31. In a log-log model the coefficient is already the elasticity, so check which form your course uses.
Does a high R-squared mean my demand estimate is reliable?
Not by itself. R-squared measures how much variation the model explains, not whether each coefficient is unbiased. If price is set partly in response to demand, for instance through discounting in slow markets, the price coefficient can be wrong even when the fit looks excellent. Graders tend to value a paper that recognizes this over one that simply reports 0.94.
Which software is expected for this unit?
Many GB540 sections use the regression tool in Excel, and some allow other packages. The tool matters less than the reporting: coefficients, standard errors or t statistics, R-squared and the number of observations. Paste raw output only if your instructions ask for it, and always add your own table and a written interpretation beneath it.
Where can I find a free GB540 Unit 3 sample paper?
This page carries a full GB540 Unit 3 demand estimation exercise, regressing electric outboard sales on four factors and converting each coefficient into an elasticity.
Why might a price coefficient be biased?
If sellers cut prices when they expect weak demand, low prices coincide with low demand, which makes buyers look less sensitive to price than they really are.