Data Science

Despite the plethora of data available to us today, data is often underutilized due to time constraints and formulaic approaches that preclude original and thoughtful data modeling. I specialize in the extraction of valid and reliable output from data - typically starting with conversations with you to clarify your needs and goals, in order to formulate recommendations and deliverables tailored to your project.

My work runs the gamut from data discovery, data evaluation, data transformation, to data analytics and visualizations grounded in statistical learning algorithms spanning both predictive and exploratory models. I routinely collaborate with subject matter experts with deep knowledge in substantive domains; and I love coaching or collaborating with clients’ in-house analysts and data scientists to bring ideas to fruition. I particularly enjoy new product development grounded in robust methodology. 

What are your data needs?

• Deep mining of existing database tables?

• Seeking guidance or support on feature engineering?

• Wish to build valid and replicable predictive models?

• Hope to integrate and analyze multiple data streams?

• Want fresh ideas on data visualization?

• Need to design and analyze A/B tests?

• Data discovery to support or refute hypotheses?

I can help with above tasks and much more.

Prior to data modeling, much patience and meticulous attention is needed on feature engineering to explore the appropriate transformations and imputations that need to be performed on pertinent variables. I can implement independently or work with your staff to execute data enhancement procedures such as multiple imputations of missing data, item analyses and transformations, optimal scaling, and data ipsatization.

My portfolio in advanced analytics and machine learning algorithms includes:

• Predictive modeling - linear and logistic regressions, multi-level/mixed models, SVM (support vector machines), decision trees, random forests, neural networks, and more.

• Segmentation analysis - k-means & hierarchical clustering, BIC-based exploratory cluster modeling, etc.

• A/B testing, factorial experiment designs, and corresponding ANOVA and choice models

• Data reduction techniques - exploratory & confirmatory factor analyses, principal component analysis (PCA)

• Market sizing and time series forecasting, multi-dimensional perceptual mapping, etc.

I also create interactive web apps for simple simulations and calculations. I keep apprised of the latest knowledge and developments in statistical learning and psychometrics to ensure that I have in my toolbox a broad spectrum of applicable techniques.

If above is in line with what you need, contact me to discuss your project needs!