STATISTICS

Statistics

Statistics

Logistic Regression Analysis

The purpose of applying logistic regression to our model is that that logistic regression is a statistical tool with improved capacity for data analysis in clinical research and epidemiology, hence its wide use. The primary objective which solves this technique is to model how it influences the probability of occurrence of an event, usually dichotomous, the presence or absence of various factors, and the value or level thereof. Also be used to estimate the probability of occurrence of each one of the possibilities of an event with more than two categories. We all know that this type of situations is approached by regression techniques. However, linear regression methodology is not applicable because now the response variable has only two values (if we focus dichotomous), such as presence / absence of hypertension. The variables used in the study to are ethnicity, anger, and efficacy, logistic regression coefficients can be used to estimate odds ratios for each of the independent variables in the model.

Since the methodology for the estimation of the logistic model is based on the use of quantitative variables, as in any other regression procedure, it is incorrect that involved qualitative variables, either nominal or ordinal. Assigning a number to each category does not solve the problem because if you have, for example, exercise variable with three possible answers: sedentary, exercise performed sporadically, you exercise frequently, and assign the values ??0, 1, 2, means to model effects that make physical exercise is often twice that it only sporadically, which does not make any sense.

Dummy variables are qualitative variable that takes the values ??0 and 1 to be included in an econometric model for the effect of high-quality features and events on the explanatory variables. In this case, the dummy variables can account for the effect not only ...