For the insurance industry, the potential to understand customers and businesses using new dimensions represented by social and other online data can unleash significant new insights from both customer behavior and risk perspective. These insights can drive insurance
automation, underwriting efficiency, and enhanced customer experience. The objective of this project is to develop and optimize the loss prediction models empowered by those insurtech innovations. This project is a real-life actuarial data science project provided by Carpe Data and COUNTRY Financial. Carpe Data is an Insurtech company that provides insurance companies with next-generation data solutions to gain a more in-depth insight into risks. COUNTRY Financial is a US insurance and financial services company that offers a range of insurance and financial products and services, including auto, home, life, commercial insurance, etc.  We will perform tasks including, but not limited to: 

• Actuarial loss modeling 
• Feature engineering using NLP 
• Unsupervised/Supervised Learning 

Supervisors: Zhiyu (Frank) Quan, Project Manager from Carpe Data, Actuarial manager from COUNTRY Financial
Graduate Supervisor: Changyue Hu

A few years ago, a new investment approach came to market—Defined Outcome Investing (DOI). DOI often promises the delivery of the upside performance of an equity asset to a certain level, with a defined downside protection amount, over a pre-established period. The first DOI ETF was just introduced in late 2018. In the following few years, DOI has become one of the most popular financial products in the market. According to Innovator, the DOI ETFs AUM (Asset Under Management) has rapidly grown to $1.7 billion in one year and nearly $4 billion in 2 years. However, the DOI strategy can be easily replicated by synthetic option product combinations. Our research team is working with a corporate partner to analyze an optimization algorithm to construct an optimized option product portfolio as a synthetic DOI strategy with better performance and greater transparency. As a result, we plan to publish a white paper to describe this methodology and model.

Supervisors: Runhuan Feng, Industry corporate partner from a Financial Technology Company
Graduate Supervisor: Yulong Wu

Machine Learning (ML) techniques have been applied broadly in the field of actuarial science and achieve fruitful results. However, to solve a practical problem with ML, the agent needs to preprocess the dataset, choose the proper ML tools, and tweak the hyperparameters of the model. All these steps are task-specific and require expert knowledge, which may render less effective ML model performance when operated by a non-ML-expert agent. To this end, AutoML is developed to facilitate the usage of ML by non-expert agents and aims to integrate the automated data processing, model selection, and model tuning into one system. This project develops two versions (Python and R) of AutoML pipeline that adapts and modifies the existing AutoML packages (for example, caret, AUTO-SKLEARN) to fit better for the actuarial ML problems. We will perform, but not limited to, the following tasks:

• Literature review on the existing AutoML system
• Understand and modify the existing AutoML codebase
• Wrap up the results in a scientific report

Supervisors: Zhiyu (Frank) Quan, Xiaochen Jing 
Graduate Supervisor: Yuxuan Li

Sales data provides businesses with valuable information on the profiling of their customers and thus has important implications on business problems, such as the improvement of customer satisfaction, retention rate, and sale efficiency. In this project, a dataset of sales records will be provided by Luyan Pharma, which has not been analyzed and utilized to its full potential for the purpose of gaining insights into the company’s operations and customers and guiding business decision-making. Luyan is a pharmaceutical company headquartered in Xiamen, Fujian, China, with operations in health product R&D, manufacturing, and distribution. As an industry leader, it is exploring new ways of improving its services provided to its customers. This is an exploratory project, and the tasks will be performed including but not limited to:

• Data wrangling
• Feature engineering using NLP
• Supervised/unsupervised learning

Supervisors: Zhiyu (Frank) Quan, Project Manager from Luyan Pharma
Graduate Supervisor: Linfeng Zhang

AXIS Capital is one of the major insurance carriers that provide cyber policies. As the world is becoming more connected than ever, systemic cyber threats pose a great risk to businesses that rely on information technology, as well as to insurers like AXIS, who provide financial protections to those businesses. As an example of systemic cyber threats, the outage of cloud services may simultaneously impact many policyholders, thus leading to a large number of claims that AXIS shall be responsible for. To improve insurers’ resilience against this risk, this project aims to explore and identify a suite of internal systemic cyber threat scenarios that together are broad enough to cover the entire threat landscape, but individually are specific enough to be used in practice on a stand-alone basis to support business decision making and risk understanding.

Furthermore, in this project, we will try to characterize each scenario for the inference of its occurrence likelihood and severity, and the results will be used to support setting capital and pricing loads, to challenge probabilistic risk modeling approaches, to monitor and manage exposure and risk trends, and may be used to set strategy.

Supervisors: Runhuan Feng, Zhiyu (Frank) Quan, Project Manager from AXIS Capital 
Graduate Supervisor: Linfeng Zhang