In one sentence
Asymptotic theory studies how estimators and test statistics behave as sample size tends to infinity, providing large-sample approximations for bias, variance, consistency, and limiting distributions.
Three core properties (one-line definitions)
Econometric asymptotics often focuses on:
[
\hat\theta \xrightarrow{p} \theta_0 \quad (\text{consistency})
]
[
\sqrt{n}(\hat\theta-\theta_0) \xrightarrow{d} \mathcal{N}(0,V) \quad (\text{asymptotic normality})
]
[
\text{Bias}(\hat\theta) = \mathbb{E}[\hat\theta]-\theta_0 \quad (\text{finite-sample bias})
]
Why large-sample approximations help
flowchart LR
A["Finite sample (unknown exact distribution)"] --> B["n grows"]
B --> C["Limiting distribution (tractable)"]
C --> D["Approximate SEs, CIs, tests"]
Background
Asymptotic theory plays a fundamental role in statistical inference within the field of economics. It focuses on analyzing the behavior of estimators and functions of estimators as the sample size approaches infinity.
Historical Context
The roots of asymptotic theory can be traced back to the early 20th century, coinciding with the formal development of statistical science. Key contributors, such as Karl Pearson and Sir Ronald Fisher, laid the foundations that linked asymptotic properties to practical statistical applications.
Definitions and Concepts
Asymptotic theory deals with the limiting properties of statistical estimators when the sample size, denoted as \(n\), approaches infinity. This includes understanding the distribution, moments, and general performance of these estimators under large sample conditions.
- Estimator: A statistic derived from sample data used to estimate an unknown parameter of the population.
- Probability Distribution: A function that represents the probabilities of all possible outcomes of a random variable.
- Law of Large Numbers: A principle stating that as the size of the sample increases, the sample mean will get closer to the population mean.
- Central Limit Theorem: In probability theory, a theorem that states that, under certain conditions, the sum of a large number of random variables is approximately normally distributed.
Quiz
### What does asymptotic theory study?
- [x] The behavior of estimators as sample size approaches infinity
- [ ] The properties of finite samples
- [ ] Small sample performance of tests
- [ ] None of the above
> **Explanation:** Asymptotic theory focuses on the properties and behaviors of estimators and their distributions as the sample size becomes infinitely large.
### True or False: Asymptotic theory only applies to normally distributed data.
- [ ] True
- [x] False
> **Explanation:** Asymptotic theory can apply to various types of distributions, not just normal distributions.
### Which of the following is a key result of asymptotic theory?
- [x] Central Limit Theorem
- [ ] Poisson Distribution
- [ ] Chi-square test
- [ ] Bayesian inference
> **Explanation:** One of the key results in asymptotic theory is the Central Limit Theorem.
### What does "convergence in distribution" mean?
- [x] The distribution of estimators approximates a certain distribution as sample size increases
- [ ] The sample mean exactly equals the population mean as sample size grows
- [ ] The sample variance decreases to zero
- [ ] The sample median equals the population median eventually
> **Explanation:** Convergence in distribution implies that the distribution of the estimator approaches a specific distribution as n tends to infinity.
### The term 'asymptotic' implies:
- [x] Approaching but not necessarily reaching a value
- [ ] Exact equality of values
- [ ] Small sample properties are fully known
- [ ] None of the above
> **Explanation:** 'Asymptotic' relates to properties or behaviors as some limit is approached but not necessarily reached.
### Which principle states that sample averages converge to expected values as sample size grows?
- [x] Law of Large Numbers
- [ ] Central Limit Theorem
- [ ] Bayesian Theorem
- [ ] Law of Small Numbers
> **Explanation:** The Law of Large Numbers states that sample averages will converge to expected values with increasing sample size.
### Who was one of the pioneers that contributed to asymptotic theory?
- [x] Andrey Kolmogorov
- [ ] Blaise Pascal
- [ ] Karl Pearson
- [ ] Ronald Fisher
> **Explanation:** Andrey Kolmogorov made significant contributions to the development of probability theory which underpins asymptotic methods.
### How does asymptotic normality benefit statistical inference?
- [x] Simplifies calculations as sample size increases
- [ ] Makes sample distribution finite
- [ ] Removes bias in estimators
- [ ] Applies only to Bayesian methods
> **Explanation:** Asymptotic normality ensures that distributions of estimators become easier to handle and infer as sample size increases.
### What concept involves deterministically narrowing the gap between sample statistics and the population parameter?
- [ ] Asymptotic bias
- [x] Consistency
- [ ] Efficiency
- [ ] Robustness
> **Explanation:** Consistency implies estimators increasingly become more accurate representations of population parameters as sample size grows.
### Which law or theorem shows that the sample mean is approximately normally distributed for large sample sizes?
- [ ] Law of Large Numbers
- [x] Central Limit Theorem
- [ ] Chebyshev's Inequality
- [ ] Markov's Inequality
> **Explanation:** The Central Limit Theorem indicates that for a large sample size, the sample mean will approximate a normal distribution.