Research & Industry Advancements Flashcards

The CAS Institute's Certified Specialist in Predictive Analytics (CSPA) credential is specifically designed for professionals who:

Answer: Apply advanced analytics in insurance without the full CAS fellowship track

The CSPA is a standalone credential for analysts and data scientists applying predictive modeling in insurance who may not be pursuing the full CAS fellowship.

In emerging research on connected autonomous vehicles (CAVs), the primary actuarial concern for personal auto liability is:

Answer: Fundamental shift in liability from driver to manufacturer and software systems

As vehicles become autonomous, accidents may involve product liability against manufacturers and software developers rather than driver negligence, reshaping coverage structures.

The actuarial concept of 'double-counting' in the context of catastrophe model output and historical loss development most commonly refers to:

Answer: Including modeled catastrophe losses in both the trend factor and the explicit CAT load

Double-counting occurs when catastrophe losses already embedded in historical data are also explicitly loaded through a separate catastrophe provision in ratemaking.

CAS research on 'nowcasting' in property casualty insurance refers to:

Answer: Using real-time or near-real-time data to estimate current-period loss activity before claims close

Nowcasting leverages current signals—early claim reports, weather data, economic indicators—to estimate losses developing in the present reporting period.

Which statistical test is most appropriate for evaluating whether a new rating variable adds statistically significant predictive power to an existing GLM in an insurance context?

Answer: Likelihood ratio test

The likelihood ratio test compares the log-likelihoods of nested GLMs with and without the new variable, yielding a chi-squared statistic for significance testing.

In CAS ratemaking research, the 'multivariate classification' approach is preferred over univariate methods primarily because it:

Answer: Isolates the independent effect of each rating variable while controlling for correlations

Multivariate methods like GLMs estimate each variable's effect while holding all others constant, avoiding the confounding bias inherent in univariate one-way analyses.

The CAS research area of 'excess of loss reinsurance pricing' primarily applies which actuarial technique to estimate the limited expected value at a given retention?

Answer: Increased limits factors (ILFs) derived from ground-up severity distributions

ILFs derived from empirical or parametric severity distributions allow actuaries to price XL layers by computing the expected value of losses truncated at different limits.