These techniques have been designed and developed in the field of medical and biological sciences but there is use is also at the same time, equally seen in the field of social and economic sciences and at some places even in engineering.
When there is a researcher in the hospital and is studying about the effectiveness of a particular treatment for a terminal disease, the primary variable of interest would be the number of days that the patient who is terminally ill would have for survival. In principle, the researcher would use standard parametric and non-parametric statistic for the description of the average survival and for the purpose of comparison of the new treatment with the existing traditional methods. But there would be certain patients in the study, who would make through the entire period of the study and there would be some patients with whom contact would be lost. Of course one would not want to exclude out those patients from the study by categorizing them as missing data. These information or data for which we have only partial information or record is called as censored observation or data.
In most of the cases, the arise of censored observations takes place whenever in the dependant variable there is a representation of time to the terminal event and there is a limited time span for the duration of the study. The occurrence of censored observations may happen in various areas of research and diverse disciplines.
Survival analysis techniques are seen getting used here and method addresses the same research question, like many of the other procedures. All the methods that are there in survival analysis handle the censored data. Survival analysis offers the regression model for the estimation of the relationship of multiple continuous variables to survival time.