For this unit’s collaborative response, you will need to construct a model of shipping costs. Assume you are tasked with developing this model for a manufacturing firm located in Kearney that ships their products to buyers throughout the United States.

This model is going to be used to estimate transportation costs only, not production costs. Therefore, determinants that only affect costs associated with the actual production process should be omitted from the model. You should consider factors which affect costs incurred from the transportation of finished product from the factory to buyers. Your model should fit the following general format: Dependent variable = average shipping cost (ASC)

How you measure this average cost is your choice. You will need to decide how to measure quantity in the formula where Avg. shipping cost = (total shipping cost/quantity).

Required independent variables:

A measure of quantity: as discussed in the lecture notes, there are many possible ways to measure quantity shipped. Feel free to improvise and select a specific good that the company produces if that helps create a measure of quantity. For example, if it helps, you could assume the company is Buckle and you are shipping jeans.

Two additional independent variables of your choice: Describe the variables in your model summary. Two indicator variables: you must include indicator variables of [rail] and [air]. Where [rail] = 0 if shipped by truck or air, and [rail] equals 1 if shipped by rail. The variable [air] = 0 if shipped by truck or rail, and [air] equals 1 if shipped by air. If we assume there are only three shipping methods (truck, rail, and air), then if both the [rail] and [air] indicator variables are 0 you must be shipping by truck. Therefore, estimated coefficients for [rail] and [air] will describe the fixed effect on average shipping costs when these methods are used instead of trucks.

You may choose to use either a linear or log form model, so your shipping cost models in regression estimation form should look like: ASC = [constant] + β1 [quantity] + β2 [ind. var. 1] + β3 [ind. var. 2] + β4 [rail] + β5 [air]

Or: ln(ASC) = [constant] + β1 [ln(quantity)] + β2 [ln(ind. var. 1)] + β3 [ln(ind. var. 2)] + β4 [rail] + β5 [air] Note: if you include an indicator variable as one or more of your additional independent variables, then you should not take the log of those variables.

For this assignment:

1) Your group should specify two different shipping costs models, with each one using a different measure of quantity. The remaining independent variables can remain the same.

2) Describe each variable in your two models, and offer a brief hypothesis of how you expect the independent variables will affect average shipping costs. Pay close attention to the specifics of how your variables are measured. Your variable descriptions and hypotheses should include some discussion of the differences between your two models. 3) Offer a justification for using company money to pay you to complete the model estimations. You can treat this like an “elevator pitch”, that is, assume you happen to get on the elevator with the CEO and you have one minute to explain why the company should pay you to do this research. This is good practice for improving your sales pitch delivery, answers to job interview questions, and general communication skills. Your answer should not be too long (about one minute or one paragraph), so choose your words wisely. Your response should address the big picture while also providing enough specifics to entice the listener to want to know more. This section of your response can really benefit from collaborative editing, feedback, and re-writing within your group. Clarity and precision of language is difficult to achieve on the first draft, so please work together on this section. Be courteous to each other, and remember that constructive criticism and feedback can be helpful and should not be intended or taken as a personal attack.