Slutsky theorem (source code)

= Slutsky theorem
{c}
{title2=$X_n\Rightarrow X,\quad Y_n\to_{\mathbb P}c$}
{wiki}

= Slutsky's theorem
{c}
{synonym}

If $X_n\xrightarrow dX$ and $Y_n\xrightarrow{\mathbb P}c$ for a deterministic constant $c$, then $(X_n,Y_n)\xrightarrow d(X,c)$. The <continuous mapping theorem> therefore gives
$$
X_n+Y_n\xrightarrow dX+c,\qquad X_nY_n\xrightarrow d cX,
$$
and $X_n/Y_n\xrightarrow dX/c$ when $c\ne0$. No <independence> between the two sequences is required. The deterministic limit matters: convergence of two separate nondegenerate marginal distributions does not determine their joint limit.

For example, if $\sqrt n(\widehat\theta-\theta)\xrightarrow dN(0,v)$ with $v>0$ and $\widehat v\xrightarrow{\mathbb P}v$, then $\sqrt n(\widehat\theta-\theta)/\sqrt{\widehat v}\xrightarrow dN(0,1)$ wherever the denominator is defined. This justifies substituting a <consistent estimator> of asymptotic <variance> when constructing a <Wald confidence interval>.