23 Bootstrap methods: Applications in econometrics

摘要: Publisher Summary This chapter reviews Efron's method called the bootstrap, and briefly mentions its relation to jackknife, with a particular emphasis on econometric applications. Bootstrap literature has made tremendous progress in solving an old statistical problem: making reliable confidence statements complicated small sample, multi-step, dependent, non-normal cases. Resampling methods provide radically new solutions several modeling problems involving interdependence, simultaneity, nonlinearity, nonstationarity, instability, nonnormality, heteroscedasticity, or missing data, Hawthorne effect, more solutions. The bootstrap handles these nonparametrically intuitively, avoiding power functions, Cramer–Rao lower bounds, bias corrections for Wald Lagrange multiplier tests, such. Many early applications of econometrics attempts alternative asymptotic standard error estimates. jackknife is also used find improved estimates errors. offers potentially valuable insight into sampling distributions, beyond simpler estimation When two tests are used, their difficult determine analytically. distribution can eliminate need tedious computations some discusses post hoc technique cleverly manipulating replications, computational aspects methods, simultaneous equation dynamic models which require special setup different from traditional bootstrap.