The client requested the advanced statistical
analysis of their portfolio, to find out patterns, similarities and insights
that may assist them in reducing the attritional loss ratio.
DataObrii thoroughly analyzed the portfolio of
the company, retrieved data insights and gained a comprehensive understanding
of the current situation. The team also utilized a deep autoencoder neural
network to detect anomalies and provided a distinct analysis of outliers.
Our recommendations for reducing the
attritional loss ratio were based on the results of tree ensemble models and
the loss-cost modelling approach. In addition, the portfolio was divided into
clusters, allowing the customer to realize all the patterns, similarities
inside, and adjust the baseline strategy with new knowledge.
RESULTS & ADVANTAGES
The customer received recommendations,
supported by statistical significance tests and sensitivity analysis including
different ways of treating large losses. The implemented solution helped him to
get insights and knowledge on identifying weak profiles, leading to the
creation of opportunities for reducing the attritional loss ratio by 3%.