Hi Ramanujam,

Could you provide more specific information about your problem? Do you have any specific mean for "statistical tech"? (e.g., variance analysis?)

Otherwise, you may first take a look at this general information:

A feature selection algorithm can be seen as the combination of a search technique for proposing new feature subsets, along with an evaluation measure which scores the different feature subsets.

The choice of evaluation metric heavily influences the algorithm, and it is these evaluation metrics which distinguish between the three main categories of feature selection algorithms: wrappers, filters and embedded methods.

Wrapper methods use a predictive model to score feature subsets.

Filter methods use a proxy measure instead of the error rate to score a feature subset.

Embedded methods: Selecting variables by optimizing directly an objective function with respect to: goodness-of-fit and (optionally) number of features. This could be achieved with algorithms like least-square regression, support-vector machines (SVM) or decision trees.

Or you can also look at a similar problem I asked and have been answered by many researchers:

https://www.researchgate.net/post/Any_one_know_how_to_solve_this_problem_similar_to_Dimensionality_Reduction

https://www.researchgate.net/post/Any_one_know_how_to_solve_this_problem_similar_to_Dimensionality_Reduction

Hi Ramanujam,

Along with t-test, F-ratio, correlation coefficients etc. you can use mRMR method of feature selection based on mutual information criterion of ranking the features proposed by H. Peng et al., IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 27, NO. 8, AUGUST 2005.

The authors have provided their work as a toolbox which is freely available for use. 

PCA is the traditional method.  I've had good luck using ANOVA (Analysis of Variance) to rank features and select the strongest to use.