skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Input-output multiplier distributions from probabilistic production paths

Thesis/Dissertation ·
OSTI ID:5686580

In the standard Leontief input-output model, a single dominant technology is assumed in the production of a particular commodity. However, in the real world, quite similar commodities are produced by firms with vastly different technologies. In addressing this limitation, the Probabilistic Production Path model (PPP) is used to investigate both the method of production and identity of the producer. An important feature of the PPP model is the consideration of the effects that heterogeneous technologies and dissimilar trade patterns have on the properties of the distribution of input-output multipliers. The derivation of the distribution of output multipliers is generalized for discrete probabilities based on market shares. Due to the complexity of the generalized solution, a simulation model is used to approximate the multiplier distribution. Results of the model show that the distributional properties of the multipliers are unpredictable, with the majority of the distributions being multimodal. Typically, the mean of the multipliers lies in a trough between two modes. Multimodal multiplier distributions were found to have a tighter symmetric interval than the corresponding standard normal confidence interval. Therefore, the use of the normal confidence interval appears to be sufficient, though overstated, for the construction of confidence intervals in the PPP model.

Research Organization:
Nebraska Univ., Lincoln (USA)
OSTI ID:
5686580
Resource Relation:
Other Information: Thesis (Ph. D.)
Country of Publication:
United States
Language:
English