Claims arrive according to a Poisson process with rate 3 per hour. What is the probability of no claims in the next 2 hours?
Mas I Probability Models
Question · Easy
Overfitting: I memorized the practice test
Short
Which statement is true for a homogeneous Poisson process?
For a homogeneous Poisson process, suppose exactly one event occurs in the interval [0,4]. What is the probability that this event occurs before time 1?
Question · Medium
Finance Exam Prep Podcast
Listen on Spotify
Podcast
A loss amount X is uniformly distributed on [0,100]. What is the limited expected value E[min(X,60)]?
A lifetime has constant hazard rate 0.02. Which survival function is correct?
A lifetime has survival function S(t) = exp(-0.02 t). What is the median lifetime?
For 0 <= x <= 100, suppose the survival function is S(x) = ((100 - x) / 100)^2. What is the hazard rate at age x?
Question · Hard
Two independent lives have one-year survival probabilities 0.96 and 0.97. What is the probability that the joint-life status survives one year?
Two independent lives have constant forces of mortality 0.02 and 0.03. If the force of interest is 0.04, what is the actuarial present value of a continuous whole life insurance paying 1 at the first death?
For a fully discrete whole life insurance and whole life annuity-due on the same life, i = 0.05 and a_double_dot_x = 12. What is A_x?
A life has a constant force of mortality 0.03. If the force of interest is 0.05, what is the actuarial present value of a continuous whole life insurance paying 1 at death?
For a fully discrete whole life model with i = 0.05, suppose A_x = 0.40. What is a_double_dot_x?
Claims arrive according to a Poisson process with rate 12 per year. Each claim is severe with probability 0.25 independently of everything else. What is the probability of exactly 2 severe claims in the next 6 months?
If X1, ..., X64 are iid with variance 16, what is the standard deviation of the sample mean?
Mas I Statistics
For an iid sample from a Poisson distribution with parameter lambda, which statistic is sufficient for lambda?
A two-sided z-test for a population mean yields z = 2.20 at significance level 0.05. What is the correct decision?
If X1, ..., X25 are iid Normal(mu, 9), what is the distribution of the sample mean Xbar?
Let N be a claim count with E[N] = 4, and let claim severities be iid with mean 1200 independent of N. What is the expected aggregate claim amount?
Let S be a compound aggregate loss with N ~ Poisson(3) and iid severities having mean 200 and variance 2500, independent of N. What is Var(S)?
If U1, ..., U5 are iid Uniform(0,1), what is P(max(U1, ..., U5) < 0.8)?
Lifetimes are iid exponential with rate lambda. Observed data are failures at times 2, 3, and 5, plus one right-censored observation at time 4. What is the maximum likelihood estimator of lambda?
Which statement best describes the sample mean as an estimator of the population mean for iid data with finite variance?
Which situation is an example of left truncation?
For iid Normal(mu, sigma^2) data with sigma^2 known, what is the maximum likelihood estimator of mu?
For iid Normal(mu, sigma^2) observations with both mu and sigma^2 unknown, which pair is a jointly sufficient statistic?
In a simple linear regression model y = beta0 + beta1 x + error, what does beta1 represent?
Mas I Extended Linear Models
Which link function is most commonly used in Poisson regression for count data?
What is the main purpose of a validation set in predictive modeling?
In a logistic regression model logit(p) = -1 + 0.7 x, by what factor do the odds change when x increases by 1?
Consider the model y = beta0 + beta1 x + beta2 I(group=B) + beta3 x I(group=B) + error. What is the slope with respect to x for group B?
In Poisson regression for claim counts with varying policy exposure, the usual way to model claim rate is to include log(exposure) as:
When comparing models fitted to the same response data, which model is preferred by AIC?
A residuals-versus-fitted plot shows a clear fan shape, with spread increasing as fitted values increase. What does this most strongly suggest?
Which display is most useful for quickly spotting outliers and skewness in a single quantitative variable?
In an ordinary least squares regression, a normal Q-Q plot of residuals shows a strong S-shape. Which assumption is most questionable?
What is the primary advantage of k-fold cross-validation over a single train-test split?
Which statement correctly distinguishes ridge regression from lasso regression?
Sensitivity in a binary classification problem is the proportion of:
Which statement best distinguishes a natural cubic spline from an ordinary cubic spline?
A logistic regression model is logit(p) = -2 + 0.5 x. What is the predicted probability when x = 3?
A Poisson regression model is log(mu) = -1 + 0.4 x + log(exposure). What is the expected count when x = 2 and exposure = 3?
If count data show substantial overdispersion relative to a Poisson model, which alternative is often more appropriate?
In a log-linear Poisson model log(mu) = beta0 + 0.08 x, how should the coefficient 0.08 be interpreted?
Why can training error alone be a poor criterion for model selection?
Consider the model y = 2 + 1.5 x + 4 D - 0.5 x D + error, where D = 1 for group B and 0 for group A. At x = 6, what is the difference in mean response between group B and group A?
A data point has a very large Cook's distance in a regression fit. Why is that concerning?
Loading more…