Premium
Statistical Aspects of Kinetic Modeling for Food Science Problems
Author(s) -
BOEKEL M.A.J.S.
Publication year - 1996
Publication title -
journal of food science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.772
H-Index - 150
eISSN - 1750-3841
pISSN - 0022-1147
DOI - 10.1111/j.1365-2621.1996.tb13138.x
Subject(s) - mathematics , least squares function approximation , arrhenius equation , linear regression , kinetic energy , exponential function , product (mathematics) , linear least squares , nonlinear regression , regression analysis , statistical model , thermodynamics , statistics , activation energy , linear model , chemistry , physics , mathematical analysis , geometry , quantum mechanics , estimator
Statistical techniques to estimate kinetic parameters (rate constants, activation energy, pre‐exponential factor) have been reviewed. Differences between non‐linear and linear regression were indicated. Extended least‐squares was shown to be useful to obtain information about experimental uncertainties of data. Measurement of reactants and products simultaneously (multiresponse) provides the possibility to estimate parameters more accurately than with uniresponse modeling (in which only one reactant or only one product is analyzed). Four examples were used to illustrate: (1) possible bias introduced by linearizing a first‐order equation; (2) use of extended least‐squares; β) advantages of multiresponse modeling; and (4) statistical problems associated with the Arrhenius equation.