- #1
maverick280857
- 1,789
- 4
Hi
I was studying the WLLN and the CLT. A form of WLLN states that if [itex]X_{n}[/itex] is a sequence of random variables, it satisfies WLLN if there exist sequences [itex]a_{n}[/itex] and [itex]b_{n}[/itex] such that [itex]b_{n}[/itex] is positive and increasing to infinity such that
[tex]\frac{S_{n}-a_{n}}{b_{n}} \rightarrow 0[/tex]
[convergence in probability and hence convergence in law] where [itex]S_{n} = \sum_{i=1}^{n}X_{i}[/itex]. For now, suppose the random variables are independent and identically distributed and also have finite variance [itex]\sigma^2[/itex].
The Lindeberg Levy Central Limit Theorem states that
[tex]\frac{S_{n}-E(S_{n})}{\sqrt{Var(S_{n})}} \rightarrow N(0,1)[/tex]
[convergence in law]
Now, if we take [itex]a_{n} = E(S_{n})[/itex] and [itex]b_{n} = \sqrt{Var(S_{n})} = \sigma\sqrt{n}[/itex], conditions of both the theorems are satisfied. But, the limiting random variables are different. In the first case, the normalized random variable tends to a random variable degenerate at 0 (in law/distribution) whereas using CLT, it tends to a Normally distributed random variable with mean 0 and variance 1.
Does this mean that convergence in law is not unique? What are the implications of these results?
Thanks.
I was studying the WLLN and the CLT. A form of WLLN states that if [itex]X_{n}[/itex] is a sequence of random variables, it satisfies WLLN if there exist sequences [itex]a_{n}[/itex] and [itex]b_{n}[/itex] such that [itex]b_{n}[/itex] is positive and increasing to infinity such that
[tex]\frac{S_{n}-a_{n}}{b_{n}} \rightarrow 0[/tex]
[convergence in probability and hence convergence in law] where [itex]S_{n} = \sum_{i=1}^{n}X_{i}[/itex]. For now, suppose the random variables are independent and identically distributed and also have finite variance [itex]\sigma^2[/itex].
The Lindeberg Levy Central Limit Theorem states that
[tex]\frac{S_{n}-E(S_{n})}{\sqrt{Var(S_{n})}} \rightarrow N(0,1)[/tex]
[convergence in law]
Now, if we take [itex]a_{n} = E(S_{n})[/itex] and [itex]b_{n} = \sqrt{Var(S_{n})} = \sigma\sqrt{n}[/itex], conditions of both the theorems are satisfied. But, the limiting random variables are different. In the first case, the normalized random variable tends to a random variable degenerate at 0 (in law/distribution) whereas using CLT, it tends to a Normally distributed random variable with mean 0 and variance 1.
Does this mean that convergence in law is not unique? What are the implications of these results?
Thanks.
Last edited: