![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Yesterday.
Oct. 2nd, 2010 02:33 pmYesterday was one of those rare days that started rotten and then just got better. It started with being woken up in the middle of the night with nightmares which I have managed to forget. Once I fell asleep again I was woken up a few hours later by yet more nightmares which involved watching a close friend die in a car accident. At that point I decided to get up.
I then soon found out that our plans for the day were shot. Now, they hadn't been important plans, but I'm getting cabin fever since all of my social events seem to keep getting themselves either postponed or canceled. This was the third one this week, and I'm starting to feel like a hermit in consequence.
However, from that point on, things started looking up. Nothing dramatic, just a bunch of small good pieces of news managed to lighten the day. My mom called (which is always a high point) and asked how are basement had fared in the torrential rains, and that's when I noticed that our backflow arrester had done its job, and we had not gotten any water in the basement. That's a relief.
Then we heard that something had gone right in a complicated financial arrangement we've been struggling with. Fingers are still crossed there, but that's something that has simply not gone right from day one, and any positive news on that front is most welcome.
Since I wasn't going to go anywhere, I decided to spend the day working on Role-Playing stuff as a proxy to actually interacting with people. As is typical for me, that involved working out some complicated statistical math which had been stumping me for the last week. Frankly, statistics is the only math course I've ever failed, although I mostly blame the lousy way its typically taught. Anyway, the intent is to come up with some tables to allow for quick calculations of barrage attacks. In other words, if 50 wizards are firing lightning at a target, how do I figure out how many hit without making 50 attack rolls? Its easy enough to figure out the average case, but I wanted to be able to generate a random selection in the distribution of possible outcomes, and do so in such a way that it can be summarized in a table for easy reference.
It did not take me long to work out that I would need to calculate an inverse binomial distribution to do this calculation. And that's were I'd been stuck for most of a week. I couldn't find a formula for it anywhere. Luckily I got help from an actual statistician, and that has put me on the right track. (It turns out there is no formula, it has to be found by numerical methods, but the C++ Boost library already contains the appropriate methods, under the name of a 'quantile(binomial(...))' ). Then, not long after, I stumbled upon a solution to a different part of that same problem, so now it seems likely that I have all I need to do the work.
So, having a bunch of things all go right sort of balanced out a day that started out really rotten.
I then soon found out that our plans for the day were shot. Now, they hadn't been important plans, but I'm getting cabin fever since all of my social events seem to keep getting themselves either postponed or canceled. This was the third one this week, and I'm starting to feel like a hermit in consequence.
However, from that point on, things started looking up. Nothing dramatic, just a bunch of small good pieces of news managed to lighten the day. My mom called (which is always a high point) and asked how are basement had fared in the torrential rains, and that's when I noticed that our backflow arrester had done its job, and we had not gotten any water in the basement. That's a relief.
Then we heard that something had gone right in a complicated financial arrangement we've been struggling with. Fingers are still crossed there, but that's something that has simply not gone right from day one, and any positive news on that front is most welcome.
Since I wasn't going to go anywhere, I decided to spend the day working on Role-Playing stuff as a proxy to actually interacting with people. As is typical for me, that involved working out some complicated statistical math which had been stumping me for the last week. Frankly, statistics is the only math course I've ever failed, although I mostly blame the lousy way its typically taught. Anyway, the intent is to come up with some tables to allow for quick calculations of barrage attacks. In other words, if 50 wizards are firing lightning at a target, how do I figure out how many hit without making 50 attack rolls? Its easy enough to figure out the average case, but I wanted to be able to generate a random selection in the distribution of possible outcomes, and do so in such a way that it can be summarized in a table for easy reference.
It did not take me long to work out that I would need to calculate an inverse binomial distribution to do this calculation. And that's were I'd been stuck for most of a week. I couldn't find a formula for it anywhere. Luckily I got help from an actual statistician, and that has put me on the right track. (It turns out there is no formula, it has to be found by numerical methods, but the C++ Boost library already contains the appropriate methods, under the name of a 'quantile(binomial(...))' ). Then, not long after, I stumbled upon a solution to a different part of that same problem, so now it seems likely that I have all I need to do the work.
So, having a bunch of things all go right sort of balanced out a day that started out really rotten.
