Statistical Modeling Techniques

Abstract

Statistical modeling data is the process of applying statistical analysis to observed data and using mathematical equations to obfuscate information derived from the data. Some statistical models can act as baseline-predictive models that help to understand advanced modeling techniques. Machine learning methods like neural networks and optimal designs can eventually provide outcomes with accurate predictions. So, it is safe to say that there is a thin line between machine learning and statistical modeling. Statistical modeling is used in a number of domains like genomics, metabolomics, and proteomics and other omics data. Statistical modeling is used extensively in many domains like pharmacogenomics, which is also called stratified health care that describes about the treatment strategies of different patient behavior genetic variability and in precision medicine. Statistical studies also showed a promising screening of drugs in drug discovery and drug development. The data collected after screening by using MS, NMR is combined with statistical techniques such as univariate analysis, multivariate analysis, PCA, variability analysis, probabilistic modeling, and support vector machines which help in decision-making process. Advantage of these techniques can help in identification of biomarkers through predictive modeling that can increase the capacity of patient’s survival and analyzing high-dimensional scaled profiling data.