Multivariate Statistical Analysis Natural And Physical...

Multivariate Statistical Analysis Statistical methodology designed to obtain information from data sets that include simultaneous measurements on many variables is called multivariate statistical analysis (1). Multivariate statistics helps to study how the variables are related to one another, and also how they work in combination to differentiate between the cases on which the observations are made. Several research disciplines such as biology, medicine, environmental Science, Psychology, Sociology, Economics, Education, Archaeology, Anthropology have broad applications of Multivariate statistical analysis (2). Numerous interesting research questions about natural and physical processes are so complex that they require multivariate†¦show more content†¦The usual regression method does not work when we are given a binary response variable. In these situations we should think about using the logistic regression (4). Considering the following model estimated regression equation (equation 1) It is called as simple Logistic regression, because there is only one predictor and also the exponential function is never â€Å"0† or negative. There are some other models available for dichotomous and non-dichotomous categorical outcomes. Probit regression: A model used for binary outcomes, but instead of the logit specification, the probit uses the cumulative distribution function for a standard normal distribution. Multinomial Logistic Regression: A model used for outcomes that are nominal, e.g., blood type (A, B, AB, O). Ordinal Logistic Regression: A model used for outcomes that are ordinal, e.g., Likert scale questionnaire (excellent, very good, good, fair, poor). Simple logistic regression application involves one dichotomous variable and one independent variable. Multiple logistic regression application involves when there is a single dichotomous outcome and more than one independent variables (4) (5). Applications in Bio-Medical Research: Logistic regression is one of the widely used tool in fields such as bio-medical research, medical, epidemiology, social sciences, engineering, ecology, psychology and marketing. For instance logistic regression can be used to