Statistics & Data Science

[Freelance Services]

Scientific data analysis has a wide variety of applications, from determining high level business or marketing strategies, to analysing statistical experiments, to building fully automated decision making systems (e.g. real-time fraud-detection, algorithmic trading etc)

We are fully literate in a wide spectrum of mathematical statistics and data scientific techniques, and are able to deliver these to a high standard, helping you:

→ gain powerful insight into your data and build predictive models

→ identify outliers or anomalous behaviour with minimum error

→ automate decision making processes using optimal decision boundaries.

In addition and if relevant to your requirements, we can then code these techniques into a real-time software framework

Machine & Deep Learning

• Supervised [ Support Vector Machines, Decision Trees, Random Forests, Artificial Neural Networks ]
• Unsupervised [ KMeans & Hierarchical Clustering, Density Clustering, PCA ]
• Time Series [ Recurrent Neural Networks ]
• Dimensionality Reduction [ LDA, PCA, TSNE ]
• Regression [ Multivariate Linear / Non-Linear / Logistic, Ridge, Lasso, Elastic Net etc ]

Mathematical Statistics

• Parametric Estimation [ MLE, Cramer-Rao, Lehmann-Scheffe, Bayesian etc ]
• Confidence Intervals [ Pivotal Quantity, Asymptotic MLE Distribution, Bayesian etc ]
• Hypothesis Testing [ UMP, GLR, Chi-Square / T / F tests, ANOVA (1-Way, 2-Way) etc ]
• Non Parametric [ Glivenki-Cantelli, Kolmogorov-Smirnov, Hypotheses tests for Quantiles, Distributions etc]