Soylent Brice
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 07:44 AM
Original message |
question for stock brokers and/or scientists |
|
i'm doing some research and i'm running into a snag. i'm aware that a lot of brokers are using genetic algorithms and neural nets for stock forecasting and predictions.
what i'm trying to find is data from some of these companies to see whether or not the Nov '08 crash was predicted by any of them, and to what extent.
if anybody has any ideas please let me know. thanks in advance!
|
Warpy
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:05 AM
Response to Original message |
1. The Royal Bank of Scotland |
|
was the only institution I know of to issue early warnings. Whether their math predicted it or they were honest enough to voice the awful realization that conservative dogma had failed them (again) is anyone's guess.
However, if you're looking for institutions relying on esoteric formulas to calculate profit and loss, I'd start there.
|
Soylent Brice
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:08 AM
Response to Reply #1 |
|
right now i'm searching the IEEE to see if i can find anything. not having any luck. and for wahtever reason, maybe my google fu is not strong this morning, i'm not having much luck anywhere else on the tubez.
i'll start there. again, thank you!!
|
Soylent Brice
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:22 AM
Response to Reply #1 |
7. found what you were talking about: |
Jim__
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:11 AM
Response to Original message |
3. I can point you at a study from 2009 asking that question - it includes neural nets. |
|
Edited on Fri Sep-25-09 08:13 AM by Jim__
You would have to purchase the study (about $32): A cross model study of corporate financial distress prediction in Taiwan: Multiple discriminant analysis, logit, probit and neural networks modelsI wouldn't expect genetic algorithms to have foreseen this. The cause of this was largely AAA rated securities that really had no value. I'm not sure this has happened before, and if it hasn't, it's unlikely that a genetic algorithm would be sensitive to it.
|
Soylent Brice
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:25 AM
Response to Reply #3 |
|
i was just reading another story about a model the Taiwanese were working on in April of last year.
i'll check this out too.
|
Kansas Wyatt
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:16 AM
Response to Original message |
4. I'm sure Web Bot predicted it... |
|
If what they claim about it is true.
|
Recursion
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:18 AM
Response to Original message |
5. Neural nets can make statistical predictions |
|
e.g., "in the next decade we can expect two 3-standard-deviation events like this crash", or whatever, not "it will happen on August 25th, 2008".
|
raccoon
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:18 AM
Response to Original message |
6. I thought they examined the entrails of animals for stock forecasting and predictions. |
|
If you ask me, they might as well.
|
BadgerKid
(1000+ posts)
Send PM |
Profile |
Ignore
|
Fri Sep-25-09 08:25 AM
Response to Original message |
|
did the companies trade on that information? If they trust their models enough to risk real money, it would make the bailout money that more curious....
|
DU
AdBot (1000+ posts) |
Thu Apr 25th 2024, 12:25 AM
Response to Original message |