AUTO INDUSTRY

Auto Industry

Auto Industry

The product acts like the prism or a transparent glass that separates light that passes through it into the colors of the rainbow. The product attributes can be the prism, the consumers' knowledge can be the light, and their response can be the different colors of the rainbow.

In this study the product includes the automobile i.e. cars. Cars were selected as the product for this study. Cars have several attributes to satisfy the consumers' needs and wants. All the attributes of cars were made to build brand equity in order to attract consumers into the brand. When the perceived quality and value of product is discussed, the general comments on the dimensions of product quality are discussed. From these criteria, this study could compare general dimensions of product with other researches.

Although stratified analysis may be used to examine confounders, this approach quickly becomes problematic when many confounders exist. Many 2 × 2tables need to be generated and analyzed, and as the number of tables grows, so does the potential for zero values in the table cells, which can lead to a poor estimate of association strength. Multivariate regression methods may be used to study these associations while taking into account all the potential confounders. Logistic, log-binomial, Poisson, and linear regression are discussed here to provide insight into these methods.

Most regression equations model the relationship between an outcome measure and a function (e.g., logit, log) of a linear combination of the independent factors and regression parameters. In studies designed to estimate the measure of association, the independent variables are made up of the exposure(s), potential confounders, and interaction terms for potential effect modifiers. The key difference in analysis between studies of prediction and studies of association is variable selection. For studies measuring associations, classic stepwise regression techniques are not appropriate; rather, variables need to be assessed with regard to their role as potential effect modifiers and confounders.

In a regression analysis, all information about association between exposure and outcome is stored in the slopes (b) of factors that contain the exposure term. Consider the following general combination of independent factors for a study:

ß0 + b1 × E

ß0 + b1 × E + b2 × C1 + b3 × C2 + b4 × C3

ß0 + b1 × E + b2 × C + b3 × µ + b4 × E × µ

ß0 + b1 × E + b2 × C1 + b3 × C2 + b4 × C3 + b5 × µ1 + b6 × µ2 + b7 × E × µ1 + ß8 × E × µ2

where ß (beta) is the regression coefficient

E is a dichotomous exposure (1 =exposed, 0 = not exposed)

Cs are potential confounders: C1 is dichotomous (1 = present, 0 = absent), C2 is dichotomous (1 = present, 0 = absent), C3 is continuous

Ms are potential effect modifiers: µ1 is dichotomous (1 = present, 0 = absent), µ2 is continuous

Typically, a model including all factors of interest is created and fit to the ...