Off-campus South Dakota State University users: To download campus access theses, please use the following link to log into our proxy server with your South Dakota State University ID and password.

Non-South Dakota State University users: Please talk to your librarian about requesting this thesis through interlibrary loan.

Document Type

Thesis - University Access Only

Award Date

2014

Degree Name

Master of Science (MS)

Department

Economics

First Advisor

Matthew Diersen

Abstract

The purpose of this research is to compare and analyze several different yield forecasting methods. The study analyzes corn yields in Ohio and South Dakota for the years 1986 through 2012. A base model, with a trend and state dummy variable is developed. Two competing models, one with objective variables and one with subjective variables, are then developed as additions to the base model. The competing objective model is developed by adding accumulated growing degree days (GDD) and accumulated rainfall variables. The competing subjective model is developed by adding a USDA crop conditions index (CCI) variable. The models are estimated weekly between weeks 24 and 36 of the calendar year. The three models are compared using several different criteria. Examinations of adjusted R2 values, F-test values, and root Mean Squared Error (MSE) values are conducted, as well as statistical tests of the competing model forecast errors. The results show that the competing subjective (CCI) model performs the best at forecasting corn yield during the growing season. It outperforms the base and objective models for the entire study period. With a minimum MSE of 8 bushels per acre, it is over 7 bushels per acre more accurate at forecasting yield than its competitors.

Library of Congress Subject Headings

Corn -- Yields -- Forecasting
Corn -- Yields -- Forecasting -- Mathematical models

Description

Includes bibliographical references (pages 64-66)

Format

application/pdf

Number of Pages

74

Publisher

South Dakota State University

Share

COinS