Colin K

@ShaneMadden Cool, thank you

I don't want the answer here, I jsut want to know if that would be on topic for the main site

Would serverfault be the right place to ask about this? I want to know if gracefully reloading my Apache configuration (via /etc/init.d/httpd reload of apachectrl -k graceful) will require me to re-enter the ssl certificate password.

Meh, can probably ignore it for now. Thanks anyway :)

Usually a subscript theta in his notation means "Conditional on theta" but that doesn't make sense in the context where I'm seeing it, and I've never seen him use it on an \equiv

This professor uses an irritating mixture of modern and ancient notation, and I don't know what the heck he means in this case :)

@DanielFischer: I don't think it's that in this case. That I've seen before. This is in the context of statistics, \theta is a model parameter

oops, hit enter early :)

@daniel

Hey has anybody ever seen the notation $\equiv_\theta$ ? That is, a triple-bar equals sign (the "is defined as" symbol), with a subscript?

That is, a triple-bar equals sign, with a subscript?

Hey guys. has anybody ever seen the notation $\equiv_\theta$ ?

I've found the PHP LDAP libraries, and we've been using the OCI8 functions for a while now, but so far we've only been connecting as an unprivileged generic user.

I can't find anything that will tell me if there is any secure way to do both operations at the same time

The user accounts in LDAP are mirrored in the oracle database, and different users already have access privileges to different database tables, and we'd like to use the same login that authenticates them against LDAP to open the database connection handle for that users session

The issue is this: We want users to be able to log in to our website securely. Right now we've just got the apache server set to authenticate users against our LDAP database, but we'd like to move to a PHP authentication in the page itself. In part of this page, the users can edit some information (like a calendar) that gets stored in an oracle database.

So, I'm not a web developer at all, but I work in a strange place, and I'm supervising an intern who is supposed to be working on a web page for us. I'm hoping you guys can point me in the right direction, or possibly help me construct a question for webmasters.se

Thanks for all your help, you guys are a great bunch as always

I've got to run for the moment, but when I get back I'll bug the sciComp people to see if they'd like the question. If so, I'll edit it and flag it for migration. If not I'll clarify some of the statistical points and leave it here, since it seems there are some aspects that interest the stats community

wow I can't spell today

If my erros were something nastier it might require a more heavy duty statustician

although finding standard errors is statistical, its a simple neough case that numerical analysis people would still understand it

I suppose this would be more statistical in nature if I didn't have reason to believe my errors were gaussian. Since I know that (and correct me if I'm wrong) the MSSE and the MLE are identical, so this becomes more about optimization than statistics

Anyway, do you think it would be best for me to edit my question and have it migrated to sciComp? Or would this be one of the rare cases where it would be best to make a more statistically flavored version for stats and a more computational flavored version for sciComp?

(for gaussian errors)

In fact it was quite dificult to convince myself that in the SSE case the covariance matrix is sigma^2 inv(H), because most people that talk about inverting the hessian are working with a MLE estimate, which is equivalent to having that factor of sigma^2 in the error function

@whuber Indeed, this is what I've done in practice. I use the J'J method because I get the last jacobian out of my fitting routine for free, while the Hessian is not easily obtained, so I'd have to compute it myself after the optimization

I'm mainly interested in understanding which estimate of H is the better estimate

nobody talks about the case of having analytical Jacobian, but NOT an analytical hessian

mainly because I can't find any resources that cover this particular case. There are lots of people saying to use J'J in the context of SSE minimization, and a lot of people saying to use inv(H) in the context of MLE, and people saying to avoid a FD Hessian when you don't have any analytical derivatives

np :)

also I've already taken great pains to avoid floating point issues/rounding error, etc

not that that's a rigorous claim :)

given the form of the derivatives it's likely that any errors would result in huge inaccuracy in the output, so I feel pretty good about it

@whuber: indeed, I can't test for that exhaustively, but the agreement with FD suggests that I haven't made any gross errors

I haven't deleted any coments

oh, not my question, nevermind

@Glen_b: My question? Thank you :)

the analytical gradients are pretty solid, and much faster than finite differences, obviously

I tested the analytical gradients against finite differences

@whuber How would I go about using that in presenting the result to others, for example? If they use the pulse area to infer other information, but the baseline is physically meaningless, should they be interested in much other than the error bars on the area?

I'd be happy to do that, depending on how much free time I find :) although the analytical gradients for this fit function are hideous

If I see significant cross covariance, is there anything I should do with that knowledge? My understanding is that it's not necessarily a bad thing as long as the fit converges

I probably won't end up implementing it in the final code, but I'dd definitely do some tests, if only to verify that it's not causing any problems

I'd see that in the off-diagonal elements of the covariance matrix, right?

hmm, that's not a bad idea

but in this case I haven't had any issues. There is so much baseline available in the data that I can subtract it out with high accuracy

that's true

although I'd considered putting a baseline parameter into the fit. I think I tested it and the results weren't significantly improved

Any error in baseline is corrected by pre-processing steps