Computer Methods for Estimating Probability of Botulinum Toxin Production in Pasteurized Process CheeseUniversity of Wisconsin--Madison, 2005 - 212 pages |
From inside the book
Results 1-3 of 14
Page 30
... response variable . The response variable can also be called the dependent variable . Note that there are only two outcomes of response variable in our case : toxin has been produced or no toxin has been produced . We use " 0 " to ...
... response variable . The response variable can also be called the dependent variable . Note that there are only two outcomes of response variable in our case : toxin has been produced or no toxin has been produced . We use " 0 " to ...
Page 39
... response variable , which is the case in our study , where the response variable has only two outcomes : toxin production or non- toxin production . In the least squares method , we need to regress the response variable In [ p / ( 1 - p ) ...
... response variable , which is the case in our study , where the response variable has only two outcomes : toxin production or non- toxin production . In the least squares method , we need to regress the response variable In [ p / ( 1 - p ) ...
Page 40
Wei Zhang. 1.5 0.5 0 Y : response variable O О O Regression line from least squares method Observed data points X : explanatory variable Figure 3.1.3.1 The least squares method is suitable to fit the ordinary linear regression model ...
Wei Zhang. 1.5 0.5 0 Y : response variable O О O Regression line from least squares method Observed data points X : explanatory variable Figure 3.1.3.1 The least squares method is suitable to fit the ordinary linear regression model ...
Contents
DATA AND DATA COLLECTION | 24 |
MODELLING AND COMPUTER METHODS | 30 |
RESULTS AND DISCUSSION | 46 |
1 other sections not shown
Common terms and phrases
added addition adjustment allows analysis applied artificial neural networks assumption binary botulinum toxin production botulism called collected combinations complete confidence interval Convergence correlation corresponding curve data points data set decision developed effects elapsed environmental conditions error estimates evaluate example experimental data Figure fixed formulation factors function given Glass data growth indicates interaction Intercept interface laboratory least squares method logistic regression model maximum likelihood method microbial microbiology microorganisms NaCl needs observed parameter pasteurized process cheese percent plot portray predictive models predictors probability probability for toxin probability models process cheese products proCHEESE provides represent requires response variable salt SAS output selected separation shown shows significant software package spores statistical step Tanaka data temperature toxic toxin assay toxin production toxin production probability tree units water activity Whole