Gambling
Broke the bank at Las Vegas, or won a packet of smokes for getting your tinkle out in class? Outrageous, heroic or plain stupid bets.
Suggested by SpankyHanky
( , Thu 7 May 2009, 13:04)
Broke the bank at Las Vegas, or won a packet of smokes for getting your tinkle out in class? Outrageous, heroic or plain stupid bets.
Suggested by SpankyHanky
( , Thu 7 May 2009, 13:04)
« Go Back
Gambling - Professional
As (some of) you know, I'm a geek.
Not just an "oh look, he's on B3TA, that makes him a geek" but a full-on, I do maths for a living and in my spare time I come up with model designs that, once I get either direct or comparative data for, I test.
Indeed one might say that I am more than homo sapiens, I am homo sapiens mathematicus statisticus. Overdramatic, moi?
My specific area of specialism is around probabilistic models using either simple Bayesians or information theory / information values to produce confidence outputs or - in simple terms - yes / no decisions about whether or not a is going to given , mainly in financial services and banking.
Academically (being a bit of a div to be honest) I worked on game theory and the headspace necessary to get ones head around this is just - astonishing.
In the last several years I've gotten into neural networks, Kohonen SOM, SVM and various other forms of taking potentially infinite numbers of inputs and thus potentially infinite correlations to come up with a value.
Unlike a Bayesian model which specifies that for a specific population there are certain actions that are common to that population (i.e., a credit score) a neural network creates a model for each card that says in simple terms "for cardholder 1234, they always buy flowers with a value of £25 on a Friday at 1300 - therefore if he buys flowers on a Friday at 1300 with a value of £25 the probability of it being not him is almost nil" - therefore we get an infinite number of models, each of which is specific to the behaviour of the given cardholder. If you're in need of an example of just *how* geeky I am, writing that gave me a wee shiver and I'm sitting here with a grin on my face.
An example of this is to be found in the form of a model that I worked closely in the design / development of shortly after finishing uni and which I'd wager the majority of B3TAns have had some contact with albeit with a level of separation. For all those of us who've gone abroad, tried to use our plastic cards and they've been declined or blocked ... erm.
At this point I look sheepish and wander off - that's my model, that is, that's deciding what the probability of it being you that's making the transaction.
Hence, if you ring your bank / card issuer before you go and tell them that you're going to country X you'll find it's not blocked because there's a wee box that gets ticked saying "this card holder is overseas" and another that says "country X".
The model isn't idiot-proof but it typically pays for itself in terms of reducing fraud losses in terms of days rather than weeks or months. It also means that cardholders feel that they're being "protected" by their banks although I'm not sure that the effect of this has ever been properly quantified.
Where was I? Oh yes. Gambling = probability.
There's a moral to this story kids and it's this: I know a lot of homo mathematicus statisticus and I don't know ANY of us who play the lottery.
( , Fri 8 May 2009, 15:53, 3 replies)
As (some of) you know, I'm a geek.
Not just an "oh look, he's on B3TA, that makes him a geek" but a full-on, I do maths for a living and in my spare time I come up with model designs that, once I get either direct or comparative data for, I test.
Indeed one might say that I am more than homo sapiens, I am homo sapiens mathematicus statisticus. Overdramatic, moi?
My specific area of specialism is around probabilistic models using either simple Bayesians or information theory / information values to produce confidence outputs or - in simple terms - yes / no decisions about whether or not a is going to given , mainly in financial services and banking.
Academically (being a bit of a div to be honest) I worked on game theory and the headspace necessary to get ones head around this is just - astonishing.
In the last several years I've gotten into neural networks, Kohonen SOM, SVM and various other forms of taking potentially infinite numbers of inputs and thus potentially infinite correlations to come up with a value.
Unlike a Bayesian model which specifies that for a specific population there are certain actions that are common to that population (i.e., a credit score) a neural network creates a model for each card that says in simple terms "for cardholder 1234, they always buy flowers with a value of £25 on a Friday at 1300 - therefore if he buys flowers on a Friday at 1300 with a value of £25 the probability of it being not him is almost nil" - therefore we get an infinite number of models, each of which is specific to the behaviour of the given cardholder. If you're in need of an example of just *how* geeky I am, writing that gave me a wee shiver and I'm sitting here with a grin on my face.
An example of this is to be found in the form of a model that I worked closely in the design / development of shortly after finishing uni and which I'd wager the majority of B3TAns have had some contact with albeit with a level of separation. For all those of us who've gone abroad, tried to use our plastic cards and they've been declined or blocked ... erm.
At this point I look sheepish and wander off - that's my model, that is, that's deciding what the probability of it being you that's making the transaction.
Hence, if you ring your bank / card issuer before you go and tell them that you're going to country X you'll find it's not blocked because there's a wee box that gets ticked saying "this card holder is overseas" and another that says "country X".
The model isn't idiot-proof but it typically pays for itself in terms of reducing fraud losses in terms of days rather than weeks or months. It also means that cardholders feel that they're being "protected" by their banks although I'm not sure that the effect of this has ever been properly quantified.
Where was I? Oh yes. Gambling = probability.
There's a moral to this story kids and it's this: I know a lot of homo mathematicus statisticus and I don't know ANY of us who play the lottery.
( , Fri 8 May 2009, 15:53, 3 replies)
Very interesting post.
I have never yet had a card refused abroad, but I often come back to an answering machine message asking me to call the bank to verify that this particular Hong Kong hotel bill is mine. How can I do that? I'm in a hotel in Hong Kong FFS!
What I don't understand is the "protecting the customer" aspect of this. So far as I can tell, if my card is used fraudulently then I am protected by the bank's guarantee and will be refunded. Preventing such fraud is therefore protecting the bank, but entirely neutral so far as my finances are concerned.
The model does work however. I never bet online and the one time a £200 online bet turned up on my credit card I had the bank on the phone to me within the hour, at 8pm on a Friday night. My card was stopped minutes later. They may be protecting themselves but they're damn efficient at it.
( , Fri 8 May 2009, 17:09, closed)
I have never yet had a card refused abroad, but I often come back to an answering machine message asking me to call the bank to verify that this particular Hong Kong hotel bill is mine. How can I do that? I'm in a hotel in Hong Kong FFS!
What I don't understand is the "protecting the customer" aspect of this. So far as I can tell, if my card is used fraudulently then I am protected by the bank's guarantee and will be refunded. Preventing such fraud is therefore protecting the bank, but entirely neutral so far as my finances are concerned.
The model does work however. I never bet online and the one time a £200 online bet turned up on my credit card I had the bank on the phone to me within the hour, at 8pm on a Friday night. My card was stopped minutes later. They may be protecting themselves but they're damn efficient at it.
( , Fri 8 May 2009, 17:09, closed)
I have heard of this
mainly as one of the people I used to work with had her surprise ruined when she got a phone call a week before her birthday asking if either her or her husband had brought two ocean cruse tickets recently.
( , Sat 9 May 2009, 2:02, closed)
mainly as one of the people I used to work with had her surprise ruined when she got a phone call a week before her birthday asking if either her or her husband had brought two ocean cruse tickets recently.
( , Sat 9 May 2009, 2:02, closed)
Ooh get you with your sexy maths talk
Talk dirty to me some more.
( , Mon 11 May 2009, 9:42, closed)
Talk dirty to me some more.
( , Mon 11 May 2009, 9:42, closed)
« Go Back