IAI Actuarial Core Principles · Actuarial Statistics · Exploratory data analysis
A researcher has data on monthly premium income of 5 branches of an Indian insurer observed every month for 24 months. How should this dataset be described?
It is panel (longitudinal) data with 120 observations. Five branches are each observed over 24 months, giving 5 times 24 branch-month records. A pure time series would track one unit and a cross-section would use one time point.
- ACross-sectional data with 5 observations
- BTime series data on a single variable
- CPanel (longitudinal) data with 120 branch-month observationsCorrect
- DCross-sectional data with 24 observations
- Categorical data because branches are labelled
Explanation
Several units (5 branches) are followed over several periods (24 months), which defines panel data. The total number of observations is 5 x 24 = 120. A single time series would follow only one branch, and a cross-section would use one time point.
Did you get it right without looking?
One question tells you little. A timed set on Exploratory data analysis shows your real accuracy, how long you take and where you lose marks.
More Exploratory data analysis questions
- A sample of 8 monthly claim counts for a small Pune motor portfolio is: 4, 7, 7, 9, 10, 12, 15, 20. What is the sample mean and the median r…
- For a sample, the variances are sx² = 16 and sy² = 25, and the sample covariance is 15. Each x is then transformed to u = 3 - 2x, while y is…
- Which feature is a genuine advantage of a stem-and-leaf plot over a histogram for a small data set of 30 observations?
- Daily premium collections (₹ lakh) over 5 days are 2, 4, 6, 8 and 30. Which statement about the outlier rule 'flag values beyond 1.5 × IQR f…
- For a sample of n = 8 paired observations, Sxx = 40, Syy = 90 and Sxy = 30. What is the sample correlation coefficient?
- A scatterplot of claim size (y) against sum insured (x) shows a clear increasing relationship, but the vertical spread of points grows as x …