- #1
LoadedAnvils
- 37
- 0
The problem:
Let [itex]\mu_{n} = \frac{1}{n}[/itex] for [itex]n \in \mathbb{N}[/itex]. Let [itex]X_{n} \; \mathtt{\sim} \; \textrm{ Poisson}\left( \lambda_{n} \right)[/itex].
Let [itex]Y_{n} = n X_{n}[/itex]. Show that [itex]Y_{n} \xrightarrow{P} 0 [/itex].
Work I've done:
I've shown that [itex]X_{n} \xrightarrow{P} 0[/itex] by showing that [itex]\mathbb{P} \left( \left| X_{n} \right| > \epsilon \right) \; \to \; 0[/itex] as [itex]n \to \infty [/itex]. (By getting [itex] \epsilon [/itex] and [itex] \delta [/itex] so that the limit definition is satisfied.)
I've also tried to show convergence in quadratic mean (which did not converge to 0) and convergence in distribution (which I could not do).
Relevant equations:
[itex]X_{n} \xrightarrow{P} X[/itex] if for every [itex] \epsilon > 0 [/itex], [itex]\mathbb{P} \left( \left| X_{n} - X \right| > \epsilon \right) \; \to \; 0[/itex] as [itex]n \to \infty [/itex].
If the distribution converges in quadratic mean, it converges in probability.
If the distribution converges in distribution and it is a point distribution then it converges in probability.
Let [itex]\mu_{n} = \frac{1}{n}[/itex] for [itex]n \in \mathbb{N}[/itex]. Let [itex]X_{n} \; \mathtt{\sim} \; \textrm{ Poisson}\left( \lambda_{n} \right)[/itex].
Let [itex]Y_{n} = n X_{n}[/itex]. Show that [itex]Y_{n} \xrightarrow{P} 0 [/itex].
Work I've done:
I've shown that [itex]X_{n} \xrightarrow{P} 0[/itex] by showing that [itex]\mathbb{P} \left( \left| X_{n} \right| > \epsilon \right) \; \to \; 0[/itex] as [itex]n \to \infty [/itex]. (By getting [itex] \epsilon [/itex] and [itex] \delta [/itex] so that the limit definition is satisfied.)
I've also tried to show convergence in quadratic mean (which did not converge to 0) and convergence in distribution (which I could not do).
Relevant equations:
[itex]X_{n} \xrightarrow{P} X[/itex] if for every [itex] \epsilon > 0 [/itex], [itex]\mathbb{P} \left( \left| X_{n} - X \right| > \epsilon \right) \; \to \; 0[/itex] as [itex]n \to \infty [/itex].
If the distribution converges in quadratic mean, it converges in probability.
If the distribution converges in distribution and it is a point distribution then it converges in probability.