Sensitivity Analysis and Uncertainty Parameter Quantification in a Regression Model: The Case of Deforestation in Tanzania

Abstract

Sep 2020, Published Oct 2020AbstractIn this paper a multiple regression model for the economic factors and policy that influence therate of deforestation in Tanzania is formulated. Sensitivity analysis for parameters of explanatoryvariables using one-at-a time and direct methods is carried out and the model is fitted by classicalleast square (LSQ) and Markov Chain Monte Carlo (MCMC) methods. Uncertainty quantificationof parameters by adaptive Markov Chain Monte Carlo methods is performed. The coefficient ofdetermination indicates that 87% of deforestation rate is explained by explanatory variablescaptured in the model. Household poverty rate is found to be the most sensitive factor todeforestation, while purchasing power is the least sensitive in both methods. Model validationindicates a good agreement between the collected data and the predicted data by the model andMarkoc Chain Monte Carlo method yielded a good sample mix. Thus, the study recommends thatsince economic activities tend to increase the rate of deforestation, then policy and decisionmakingprocesses should link the country’s desire for economic growth and environmentalmanagement.
 Keywords: deforestation; economic factors; Markov Chain Monte Carlo methods; regressionmodel; sensitivity;