Evaluating Consumer Usage of Nutritional Labeling: The Influence of Socio-Economic Characteristics

Where:

F(Zi) = represents the value of the standard normal density function
associated with each possible value of the underlying index Zi.

Pi    = the probability that an individual is a frequent user of nutritional

labeling given knowledge of the independent variables Xis

e    = the base of natural logarithms approximately equal to 2.7182

Zi    = the underlying index number or βX_i

α = the intercept

And βX_i is a linear combination of independent variables so that:

Zi= log [Pi∕(1-Pi)] = β о + β 1X1 +β 2X2 + . . . +β nXn + ε

Where:

i = 1,2,. . . ,n are observations

Zi    = the unobserved index level or the log odds of choice for the i^th

observation

Xn    = the n^th explanatory variable for the i^th observation

β    = the parameters to be estimated

ε     = the error or disturbance term

The dependent variable Zi in the above equation is the logarithm of the probability that
a particular choice will be made. The parameter estimates do not directly represent the
effect of the independent variables. To obtain the estimators for continuous
explanatory variables in the logit model, the changes in probability that Yi = 1(Pi)
brought about by a change in the independent variable, Xij is given by:

(∂Pi/ ∂Xij) = [β j exp (-β Xij)] / [1+exp (-β Xij)] ²

<<   7   8   9   10   11   12   13   >>

More intriguing information