Modeling & Statistical Analysis: Get the most out of your data
Our analysts are trained in many types of research designs. In analyzing data, we use the most advanced statistical and mathematical applications, including time-series forecasting, generalized least square econometrics, logit and probit categorical dependent variable analysis, survival modeling, causal path modeling, and analytical hierarchy process.
In addition, we have developed many in-house computer programming routines for tackling various specialized research topics. Listed below are some of the data-mining methodologies employed by our firm.
Our program evaluation studies accomplish or support several objectives:
Causal Path Modeling (CPM) is a comprehensive approach that can be used to analyze multiple linkages between variables, even those variables that are influential yet not directly observable. CPM is an advanced statistical tool based on structural equation modeling technique.
Neural network (artificial intelligence) programming is used to sift through data for patterns that will predict behavior and uncover hidden, often non-linear, patterns.
Cluster analysis is used to group similar people together based on similar data characteristics.
Discriminant analysis is a classification method that measures the importance of factors determining membership within a defined category.
Decision Trees utilizing CHAID and CART algorithms separate out data into sets of rules which are likely to have a different effect on a target variable.
To learn more, contact us.