Document Type

Master's Culminating Experience

Publication Date

1997

Abstract

The objective of this paper is to determine the feasibility of using hedonic pricing models to predict the cost to the Air Force of aircraft engines purchased under government contracts. A hedonic pricing model is constructed using an ordinary least square regression format. The distinguishing factor is that cost is always the dependent variable in a hedonic pricing model. The word hedonic means pertaining to pleasure. Economists relate pleasure to value. Therefore, the predicted coefficients of variables produced by the hedonic model, represent marginal values of each variable. The coefficients also show how the variables impact cost. A good hedonic model will be able to explain a large portion of the total variation in the model and accurately predict cost. A sample of twenty turbofan engines was used to construct the models. Internal engine parameters were too highly correlated to produce significant results. Four internal variables were chosen to represent the entire bundle of internal engine characteristics. These four variables, overall pressure ratio, thrust-to-weight ratio, thrust, and weight, along with four external variables were placed in various models to determine if any combination of variables would produce a model that could predict the cost paid by the Air Force of an aircraft engine. Multiple series of models were run with various levels of success. The one problem persisted through all the models--the sign associated with the variable, cost of labor (LBR), was negatively correlated with cost suggesting that as cost of labor increased cost of production decreased Tie logical progression of this is that if labor is paid more and more, the engines will cost next to nothing. Obviously, this cannot be true. Many iterations failed to correct this error. The two main reasons that this approach did not work are the small sample size and the quasi-market where government aircraft engines are purchased by the Air Force. Multicollinearity played a large part in the small sample size and political factors greatly influenced the market for government-purchase aircraft engines. For these reasons, hedonic pricing techniques could not be used to predict the cost to the Air Force of aircraft engines purchased under government contracts.

Download


Share

COinS