Remove 2007 Remove Modeling Remove Uncertainty
article thumbnail

AI-Enabled Contingency Planning is More Accessible Than Ever

David Menninger's Analyst Perspectives

We live in a time of uncertainty, not unpredictability. The ability to rapidly model and plan different scenarios at a useful level of detail enables organizations to assess alternate options more frequently and pivot quickly when conditions change enough to warrant it. Or, at least enduring the least amount of damage.

article thumbnail

Why model calibration matters and how to achieve it

The Unofficial Google Data Science Blog

by LEE RICHARDSON & TAYLOR POSPISIL Calibrated models make probabilistic predictions that match real world probabilities. While calibration seems like a straightforward and perhaps trivial property, miscalibrated models are actually quite common. Why calibration matters What are the consequences of miscalibrated models?

Modeling 122
Insiders

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

article thumbnail

Transforming FSI in ASEAN with Cloud Analytics

CIO Business Intelligence

auxmoney began as a peer-to-peer lender in 2007, with the mission of improving access to credit and promoting financial inclusion. Right from the start, auxmoney leveraged cloud-enabled analytics for its unique risk models and digital processes to further its mission. We see this demonstrated in S-Bank , ranked No.

article thumbnail

The Lean Analytics Cycle: Metrics > Hypothesis > Experiment > Act

Occam's Razor

Let's listen in as Alistair discusses the lean analytics model… The Lean Analytics Cycle is a simple, four-step process that shows you how to improve a part of your business. Another way to find the metric you want to change is to look at your business model. The business model also tells you what the metric should be.

Metrics 157
article thumbnail

Towards optimal experimentation in online systems

The Unofficial Google Data Science Blog

Crucially, it takes into account the uncertainty inherent in our experiments. Experiments, Parameters and Models At Youtube, the relationships between system parameters and metrics often seem simple — straight-line models sometimes fit our data well. It is a big picture approach, worthy of your consideration.

article thumbnail

Changing assignment weights with time-based confounders

The Unofficial Google Data Science Blog

For this reason we don’t report uncertainty measures or statistical significance in the results of the simulation. In practice, one may want to use more complex models to make these estimates. For example, one may want to use a model that can pool the epoch estimates with each other via hierarchical modeling (a.k.a.

article thumbnail

Estimating causal effects using geo experiments

The Unofficial Google Data Science Blog

On the other hand, for the models that we use, the standard error of the iROAS estimate is inversely proportional to the ad spend difference in the treatment group. The model regresses the outcomes $y_{1,i}$ on the incremental change in ad spend $delta_i$. The noisier the data, the higher the standard error. For example, $beta_2 = 3.1$