Exam SRM (Statistics for Risk Modeling) Practice Test
Gana confianza para Exam SRM (Statistics for Risk Modeling). Practica los conceptos, comprende las respuestas y refuerza tus conocimientos pregunta a pregunta.
Probar una preguntaDescripción y detalles del examen
The Society of Actuaries (SOA) Exam SRM: Statistics for Risk Modeling is a foundational professional certification for actuaries and quantitative risk professionals. This exam validates a candidate's mastery of statistical methods essential for modeling and analyzing risk in insurance, finance, and related fields. It covers the critical intersection of predictive analytics, regression modeling, time series analysis, and decision trees, providing the statistical backbone for advanced actuarial work. Earning a passing grade on Exam SRM is a mandatory step toward Associate of the Society of Actuaries (ASA) credentialing and demonstrates a rigorous, applied understanding of how statistical theory informs real-world risk assessment. Professionals who pass this exam signal to employers a proven competency in building, validating, and interpreting the statistical models that drive pricing, reserving, and enterprise risk management decisions.
Preguntas de Muestra
Elige una respuesta y consulta la explicación para ver cómo funciona la práctica.
SOA SRM 2026 case file 025: A binary lapse model produces fitted probabilities outside the interval [0,1] when an identity-link linear probability model is used. Which GLM setup most directly prevents this while modeling a binary response?
SOA SRM 2026 case file 021: For a new policy, an OLS model estimates expected annual severity at 1,200. The analyst reports both a 95% confidence interval for the mean response at that predictor profile and a 95% prediction interval for one future policy. Which interval should be wider?
SOA SRM 2026 case file 004: The reinsurer portfolio 4 fits several candidate models for ceded loss severity for cohort 4. The validation set has been used repeatedly to choose transformations, tune hyperparameters, and remove predictors. A manager asks for one number to report as the expected performance of the finalized process on new business. Which dataset should supply that number?
SOA SRM 2026 case file 005: A risk committee receives a proposal to segment 2025 credit bureau pull extract 5 without using claim outcomes, then separately profile each segment's average retail default indicator for cohort 5. Which description best classifies the first modeling step?
SOA SRM 2026 case file 120: A random forest uses a different random subset of candidate predictors at each split. Why can this improve performance relative to ordinary bagging when several strong predictors dominate?
Oportunidades profesionales y salario
Los rangos son cifras del mercado de EE. UU. salvo que se muestre un rango local.