Answer: II and III

**Correct Answer: C (II and III)** **Explanation:** **Statement I is incorrect:** The Spearman and Kendall models are actually **ordinal** correlation measures, not cardinal. They work with ranks rather than actual numerical values, which makes them **less** sensitive to outliers, not more sensitive. Cardinal observations refer to actual numerical values with meaningful intervals. **Statement II is correct:** Both Spearman's rank correlation and Kendall's tau are **ordinal correlation measures**. They consider the order/rank of observations while ignoring the actual numerical values. This is a limitation when applying them to financial data where the actual magnitude of values matters. **Statement III is correct:** Kendall's tau works by comparing pairs of observations. When there are many tied pairs (neither concordant nor discordant), these pairs are omitted from the calculation. Working with only a few concordant and discordant pairs can distort the Kendall tau coefficient, making it less reliable. **Statement IV is incorrect:** The Pearson correlation coefficient measures **linear** relationships, not nonlinear relationships. Financial variables often exhibit nonlinear relationships, which is why Pearson correlation may not be the best model for financial applications. Additionally, Pearson correlation is sensitive to outliers. **Why this matters for financial risk management:** - Financial data often contains outliers and exhibits nonlinear relationships - Ordinal correlations (Spearman, Kendall) ignore the magnitude of price movements, which is crucial in finance - The limitations of these statistical correlation models highlight the need for more sophisticated approaches like copulas or tail dependence measures in financial risk modeling

Author: Nikitesh Somanthe