B.4 Continuous Changes Flashcards
Reason for adjusting for continuous changes
To ensure that historical data reflects the mix of business and levels of social and economic inflation expected in the future period.
Possible data sources for trends
-Historical insurer data
-Industry data
-Economic data
Data adjustments before trending
Adjust for large events & anomalies, one-time changes, and seasonality.
Advantage/Disadvantage of using written premium
to determine premium trends
The advantage of using average earned premium to
determine the trend is that we are trending earned premium, so it makes sense to determine the trend on earned premium.
The advantage of using written premium is that we can use more recent data, as changes in average written premium will ultimately show up as changes in average earned premium.
Why frequency and severity are typically analyzed
separately
Since they can change for different reasons, and combining them might make the changes less apparent.
Advantage/Disadvantage of using calendar period
data to determine loss trends
The advantage of using calendar period data is that you don’t need to estimate ultimate losses, which are uncertain.
The advantage of using accident year ultimate loss data is that it provides a better match between losses and exposures.
Advantage/Disadvantage of using paid loss data
compared to reported loss data to determine loss
trends
The advantage of using paid loss data is that it is not subject to changes in case reserving practices.
The advantage of using reported loss data is that it
incorporates more recent information, since case reserves provide information you might expect to eventually see in paid data.
Two other alternative methods for loss trends
- Using econometric or generalized linear models
2. Using incremental calendar year data by accident year
Three ways to calculate a trend from historical data
- Take an average of the percent changes.
- Fit a line to the data: This assumes a constant change for each time period.
- Fit an exponential curve to the data: This assumes a
constant percent change between each time period.
Two reasons why excess severity trends are greater
than basic or total limits trends
- For losses above the basic limits, the trend is entirely in the excess layer.
- Losses just under the basic limit are pushed into the excess layer by the trend, creating new losses for the excess layer.
Two assumptions commonly made in determining
trend periods
- Policies are written uniformly over time.
2. Premiums are earned uniformly and losses occur uniformly over the policy period.
Information needed to determine the average earned
date of a future policy period
- Future rate change effective date
- Length of time rates are expected to be in effect
- Policy term length
Two ways to perform the first step of a two-step trend
- Adjust the historical period level to be equal to the latest level: This assumes the latest level is a reliable number from which to base the future forecast. This is more common for premium trends since they have less random fluctuation than loss trends. For example, if you use written premium trends to forecast earned premium, the current trend factor is:
Current trend factor = Latest Average WP at Current/
Historical Average EP at Current - Trend the historical period level to the latest time period: This is performed like a one-step trend, and is more common for loss trends.
Overlap fallacy between development and trending
Development brings the data from each historical period to its ultimate level, while trending reflects the difference in ultimate levels from one historical period to the next. In
ratemaking terms for losses, loss development makes sure that the future policy is priced to cover ultimate losses, while loss trend makes sure that the ultimate losses are at the cost levels corresponding to the future policy period. Historical losses are trended from the average accident date of the historical period to the average accident date of the future period, and loss development takes losses from the average accident date of the future period to their ultimate level.
