The ability to estimate quantiles is key to estimating coherent risk measures. Quantiles are values that divide a distribution into subsets of equal size. For example, the median is the value that divides a distribution in half. The Value at Risk (VaR) is a risk measure quantile that measures the maximum loss that can occur with a given probability. Estimating VaR necessitates the estimation of the appropriate quantile of the distribution of the portfolio's returns. Therefore, the ability to estimate quantiles is crucial for estimating coherent risk measures.

Choice A is incorrect. While the ability to assign weights to assets in a portfolio accurately is important in portfolio management, it is not specifically key to estimating coherent risk measures. Coherent risk measures are more concerned with the overall risk of a portfolio rather than individual asset weights.

Choice B is incorrect. The ability to calculate exponential value accurately does not directly relate to estimating coherent risk measures. Exponential values may be used in certain calculations within financial modeling, but they are not a primary factor in determining coherent risk measures.

Choice D is incorrect. Although approximating risk exposure can be part of the process when managing financial risks, it's not the key ability for estimating coherent risk measures which focus on potential losses and their probabilities.