P-Category for Data Science Portfolios

The purpose of the study was to identify customers at risk of leaving. The research question being tested was: Research Question 1: What is the probability that customers will leave? We built predictive models. We performed churn analyses for the utility industry by analyzing the data including addresses, gas consumption, products, contract terms and [...]

AMSTAT Consulting has applied an effective analytical approach to evaluate the cost of insurance fraud and to quantify the value of investments in counter fraud analytics capabilities in order to reduce our client's risk exposure. By incorporating this empirical data into its simple Monte Carlo model, AMSTAT’s analysis shows that the annualized cost of [...]

The purpose of this study was to predict the success of advertising campaigns by analyzing numerous data sources to determine the ROI of an advertising campaign. (e.g., Reach, Influence). We predicted the success of a particular advertising campaign based on brand, talent, external factors, and trends.  We built predictive models. There is a 90% probability [...]

The purpose of the study was to forecast sales of new sites based on the data and sales of the operating companies. The hypothesis being tested was: H1: The investments in new companies are getting a high return. We built predictive models. We used location analytics. There is a 72.35% probability that the investments [...]

The Problem Our client has to actively seek friendly fraud (aka chargeback fraud) risk. Friendly fraud occurs when an individual makes a purchase online via their credit or debit card then requests a chargeback from the bank once the goods or services have been consumed. A completed chargeback cancels the original transaction and refunds [...]

Problem The universal problem is how to quickly determine the root cause of incidents and then contain and remediate them. Once this is completed, the goal is to return intelligence from the analysis back into the system for proactive diagnostics and mitigation for continuous cybersecurity improvement Purpose of the Study The purpose of the [...]

The purpose of the study was to develop an understanding of the relevant online conversation, both pre- and post-FDA approval of [competitor 1] and to identify stakeholders and discover discussion patterns within the set of conversations. This in-depth “competitive analysis” focused primarily on conversations with patients and caregivers. The data included posts from Twitter, [...]

The purpose of the study was to measure if the phone call data that our clients capture can be used to determine predictive behavior. The hypothesis being tested was: H1: The phone call data that we capture (every interaction) can be used to determine predictive behavior. We used Natural Language Processing. We found that [...]

The purpose of the study was to predict the sales per day for each of the next 100 days. We created a general purpose data science module in python that could Digest the daily sales data for the past 3 years and develop a model Identify and use any publicly available sources of data [...]

The purpose of the study was to predict the likelihood of a claim being part of an accident. We developed the algorithm that could leverage features, both from the claims and potentially beyond the claims. We inserted an algorithm into our client's existing ETL process and provided a single, key metric:  likelihood of this [...]