This preview shows page 1. Sign up to view the full content.
Unformatted text preview: . The strength of the dependence is sometimes of
primary importance. For example, researchers in medicine are often interested in the correlation
between some behavior and some aspect of health, such as between smoking and heart attacks.
Sometimes one random variable is observed and we wish to estimate another. For example, we
may be interested in predicting the performance of a ﬁnancial market based on observation of
an employment statistic. Some questions that arise in this context are the following: What does
it mean for a predictor or estimator to be the best? What do we need to know to compute an
estimator? What computations are required and how complex are they? What if we restrict
attention to linear estimators? Is there a generalization of the Gaussian distribution, central limit
theorem and Gaussian approximation for multiple dependent random variables? 4.1 Joint cumulative distribution functions Recall that any random variable has a CDF, but pmfs (for discrete-type random variab...
View Full Document
- Spring '08
- The Land