Data Science & Predictive Analytics

Models built to be checked, not just fitted.

Prediction and inference are different problems, and conflating them is how models end up confidently wrong. DeepTeal’s work here rests on formal training in biostatistics and mathematics, which mostly shows up as scepticism: about leakage, about validation that flatters the model, and about results that look too good to be true.

Two models fitted to the same observations. The dark one is flexible enough to pass through every single point, so its error on the data it was given is exactly zero. The hollow points arrive afterwards, and neither model has seen them. The perfect score does not survive them; the duller one barely moves.

Areas of work

Forecasting and predictive models Demand and revenue forecasting, churn and retention models, and segmentation and scoring, with validation designed to survive contact with new data.
Experiment design A/B and multi-variant tests: sample size and power, what is being controlled for, how multiple comparisons are handled, and what the result can and cannot be used to claim.
Validation Checking that a model still holds: validation against data it has not seen, watching for drift, and saying when a simpler method would do the job.

For enquiries about work of this kind, get in touch.