Forecasting Turkey imposing no-fly zone on Syrian Air Force

Deep_Web_Guy

New Member
The Turkish-Syrian relationship is tricky. Turkey has been rather reserved when it comes to its southern neighbor.
But, now there are several indicators hinting that Turkey may impose a no-fly zone over Syria, specifically on the Syrian Air Force. The Turk certainly has the dominant air assets to do so, however the Syrian air defenses are stout.
I've used standard Bayesian statistics to make my prediction. If anyone proposes better updates to the probabilities of any factors, I'd like to hear it.
The bottom line prediction is 60% as of 13OCT12.
My Bayesian math is over at deep-web(dot)org in the blog under Turkey no fly prediction. Can't seem to post the link.
 

Deep_Web_Guy

New Member
  • Thread Starter Thread Starter
  • #2
With some feedback from other sources, additional factors have been added that lower a Turkish No-Fly zone down to 20%

Full article at deep-web dot org in the blog on Turkey No-Fly Zone

Bayesian math is below along with other factors. AGain, if anyone has further factors or proposed adjustments, just let me know.
+++++++++++++++
Update 16OCT12, yet again

I've been given lots of feedback to take in new factors. Main flaw is that I have ignored negative factors, so I will update my Likelihood factor.

Last left at L=5.858
Additional Factors

Syrian Air Defences are superior and will require neutralizing: P(e|h) = .20 & P(e|nh)=.40
Political Will by Turkey to get involved in Syria : P(e|h) = .4 & P(e|nh)=.6

new R= 0.11 x 5.858x .12/.4 x .4/.6

R=0.21479

Percentage = .21470/(1+.21479) = 17%

With 1 sig fig it rounds to 20% which I think is more believable.
Chime in if you think this is right.
 

swerve

Super Moderator
This is utter nonsense. GIGO. The input values are subjective impressions, or plucked from the air. Any output value derived from them is no better than a gut feeling.
 

RobWilliams

Super Moderator
Staff member
This is utter nonsense. GIGO. The input values are subjective impressions, or plucked from the air. Any output value derived from them is no better than a gut feeling.
That is the feeling i'm getting too.

Personally, I feel this would be better 'graded' using literacy; i.e Very likely, likely etc or even better - writing down points and using evidence and proof to explain points and then ultimately write down a gut instinct value, rather than this pantomime.

Seeming as the OP isn't able to post links, i'll link it for him. I have to say

Turkey No Fly Zone | Deep Web Search - A How-To Site

Seems like the biggest crock of crap i've ever seen, all the values he's using of probabilities of things happening are his guesses, no info how he got to those numbers they just popped into existance and conveniently match up almost exactly to his original "gut instinct".
 

Deep_Web_Guy

New Member
  • Thread Starter Thread Starter
  • #5
This is utter nonsense. GIGO. The input values are subjective impressions, or plucked from the air. Any output value derived from them is no better than a gut feeling.
Absolutely. This math is very susceptible to GIGO. But Bayesian stat is a valid method to convert the level of belief of an area expert into a number. I'm not claiming to be an area expert at all. Just trying to approximate a general idea of R, and identify my biggest errs.

Single analysts, be it commercial or military, are more at danger of GIGO without someone challenging the P(h) & P(nh) of the factors. Which is the case here. It is also the reason to solicit outside input.
 

Blackshoe

Defense Professional
Verified Defense Pro
I seem to recall Robert MacNamara was pretty good at using math and numbers to forecast and fight a war, too. How'd that work?
 

swerve

Super Moderator
Absolutely. This math is very susceptible to GIGO. But Bayesian stat is a valid method to convert the level of belief of an area expert into a number. I'm not claiming to be an area expert at all. Just trying to approximate a general idea of R, and identify my biggest errs.

Single analysts, be it commercial or military, are more at danger of GIGO without someone challenging the P(h) & P(nh) of the factors. Which is the case here. It is also the reason to solicit outside input.
OK, but the problem here is that you're not going to get those numbers precise enough to be useful without information that nobody who's going to respond to you on the internet will give you.

The best information which anyone has, when put into your formulae, will have enormous margins of error, & the people who have the best information won't reveal it, or they'd go to prison, at best.

You're doing the equivalent of adding two numbers both of which are between 0 & 100 - and that's all you know about them. The answer is meaningless.

If you find the mathematical analysis interesting, & are good at it, then I suggest that you apply your interest & abilities more productively. There are many problems which you could analyse where the results could be of value, or at least interest. Economic historians, for example, are always looking for ways to squeeze more information out of data. But to get information, you have to have data! The three most important things you need for your analyses to be worthwhile are data, data & data - and that's what you lack.
 
Top