188,162 reputation
17131291
bio website
location
age
visits member for 3 years, 9 months
seen 28 mins ago

As somebody used to say:

Does research. Smokes. Battles administration. Smokes. Wishes he could stop battling administration so that he could have more time to do research. Smokes some more.

The same. Except I do not smoke.


How to ask a good question?

This paragraph is for my personal use but freely available:

Welcome to Math.SE! Please, consider updating your question to include what you have tried and where you are getting stuck. That way, people on this site will know exactly what help you need.


25m
comment CI for the expected value of the sum of two dependent normal RVs
Time for some fact checking on your part, it seems... Which sources do you use on normal random variables?
27m
comment $E[e_te_s\Delta B_t\Delta B_s]$ for $\Delta B_t$ Brownian motion increments and $e_t(\omega)$ a measurable function.
This is part of the very definition of Brownian motion.
1h
answered Find $S=\frac{1}{2}+\frac{3}{2^2}+\frac{5}{2^3}+\frac{7}{2^4}+…+\frac{2n-1}{2^n}+…$
1h
comment Find $S=\frac{1}{2}+\frac{3}{2^2}+\frac{5}{2^3}+\frac{7}{2^4}+…+\frac{2n-1}{2^n}+…$
There are much less convoluted ways to compute the full sum S.
1h
comment CI for the expected value of the sum of two dependent normal RVs
onion: The random variable Y is normal with known mean E(Y)=w1mu1+w2mu2 and with known variance. Could you explain what is your question exactly?
2h
comment compare $cov(aX, bY)=ab \;cov(X, Y) $to $Var(abX)$ using the marginal distribution $f_X(x)$
+1. Which shows that the PDF $f_X$ is offtopic to prove this.
2h
comment Closed sets with empty interior measure zero
Hint: Fat Cantor sets.
2h
comment If $f$ is continuous, then $\lim\limits_{n \rightarrow \infty} \int^b_a n(f(x+ 1/n)-f(x)) \lambda(dx) = f(b)-f(a)$
It helps because then $|f|$ is b-----d.
2h
comment To find the limit of $\frac{1}{\sin n}+\frac{1}{\cos n}$
Somewhat shortened the thing, getting rid of unecessary arguments. If you object to the modifications, just revert to the previous version.
2h
revised To find the limit of $\frac{1}{\sin n}+\frac{1}{\cos n}$
deleted 244 characters in body
2h
answered Can You Pass Nonlinear Functions of Conditioned Variable Through Conditional Expectation?
2h
comment Can You Pass Nonlinear Functions of Conditioned Variable Through Conditional Expectation?
@Irvan E(eX⋅Y|X) has only one meaning.
2h
comment Probability conundrum
This is not the probability of a 5-tie.
2h
comment Joint multivariate normal distribution
"EDIT2: a must be nonzero" No.
2h
comment Joint multivariate normal distribution
@StefanHansen Every affine transform of every normal distribution is normal. For example, if y is one-dimensional normal, then (y,ay) is bivariate normal for every a, zero or nonzero. The only thing is that affine transforms of a normal distribution with density can be concentrated on a strict affine subspace, that is, be normal distributions without density.
4h
comment If $f$ is continuous, then $\lim\limits_{n \rightarrow \infty} \int^b_a n(f(x+ 1/n)-f(x)) \lambda(dx) = f(b)-f(a)$
This is almost surely not true. You know that $f$ is continuous, in particular, on the interval $[c,c+1]$, $|f|$ is $______$, hence ...
4h
comment Conditional expectation knowing $X$ and knowing $f(X)$
Proof: For every measurable function $h$ and random variable $U$, $\sigma(h(U))\subseteq\sigma(U)$. Apply this twice.
12h
comment Direct way to solve $y' = y^2$
Quote: "You might want to tell us more about this "ultimate goal", I suspect the approach in this question will not yield this result---which might actually be false."
12h
comment martingales, almost sure convergence
It <ould be interesting (and helpful) that you describe precisely how you tried using the strong LLN.
12h
comment Prove that f is uniformly continuous on [0, ∞].
"hence your function is uniformly continuous on [X,∞)." This should be reformulated: the function satisfies the property asked for uniform continuity on [X,∞) for this value of epsilon.