Microsimulating farm business performance
Kokic, P., Chambers, R. and Beare, S. (1997) Microsimulating farm business performance. University of Southampton, Department of Social Statistics Technical Report, (3), 11.
Full text not available from this repository.
Microsimulation of business performance based on sample survey data is a relatively underdeveloped field, but its application in government economic policy formulation is potentially great since it can be used to measure the distributional effects of change rather than just average change. Techniques which account for the dynamic response of businesses to macro level price expectations have recently been developed (Kokic et al., 1993). These allow individual level business performance to be forecast from sample survey data. In this paper we outline a general methodology for combining these forecasting techniques with Monte Carlo simulation in order to produce a microsimulation of business performance that accurately captures the true distributional characteristics of the underling survey data. Applying this methodology to Australian farm survey data, we show that these methods may be used to forecast the distribution of farm business production and performance within arbitrary subdomains of the surveyed population conditional on a given set of expected commodity price outcomes. The microsimulations reflect both the uncertainty due to climatic variation from one year to the next, which in the Australian context depends largely on geographic location, as well as the uncertainty of commodity prices.
|Keywords:||expectile regression, kernel smoothing, M-quantile regression, performance measurement, supply modelling|
|Subjects:||H Social Sciences > H Social Sciences (General)|
|Divisions:||University Structure - Pre August 2011 > School of Social Sciences > Social Statistics
|Date Deposited:||21 Aug 2008|
|Last Modified:||02 Mar 2012 12:05|
|Contributors:||Kokic, P. (Author)
Chambers, R. (Author)
Beare, S. (Author)
|RDF:||RDF+N-Triples, RDF+N3, RDF+XML, Browse.|
Actions (login required)