# TED Community » Richard Morris

• TEDCred score: +0.40 TEDCred reflects your contribution to the TED community.

• #### A comment onTalk: Didier Sornette: How we can predict the next financial crisis

Jun 19 2013: So trying to unpick the maths behind this. It seems that they are using a Log-Periodic Power Law. The key part of the model is (t-t0)^k(A+B cos(C log(t0-t)). The (t-t0)^k is the power rule part: prices will tend to go up and the rate of increase increases. The cos part gives some periodic variation, a bit like water waves, in the general upward trend sometimes it will be higher and sometimes lower. The log part basically means that the periods decrease. As you get closer to a bubble busts the price fluctuates faster. A nice review of this can be found here http://quantivity.wordpress.com/2011/02/08/curiosity-of-lppl/
• +2

#### A reply onTalk: Didier Sornette: How we can predict the next financial crisis

Jun 18 2013: There may be a good statistical reason for not fitting a quadratic. If you look at the bottom right you will notice there are few data points, these are the rare events of large magnitude. They are also quite widely spaced so there is a lot of uncertainty in the distribution. Yes you could fit a quadratic but there is a danger of overfitting, which might indicate false confidence.

Actually reading Sornette paper http://arxiv.org/pdf/0907.4290v1.pdf he does go into great lengths about these tails where the data differers from a power law. It is this bit which is the Dragon King
• +2

#### A comment onTalk: Didier Sornette: How we can predict the next financial crisis

Jun 17 2013: It nice to see ideas from catastrophy theory, dynamical systems and bifurcation theory are finally being taken up by the economic community. The core mathematical idea behind some of the talk goes back to René Thom in the 1960s who identified the elementary catastrophes http://en.wikipedia.org/wiki/Catastrophe_theory The slide at 15:00 shows a classic cusp catastrophy, which show how a system can rapidly switch from one state to another, the regimes mentioned in the talk. This is a universal phenomena and can be observed in many domains from stability of ships to hysteresis in magnetism. The smart bit is knowing where the folds which mark the tipping points are.
• #### A reply onConversation: Campaign to get the UN to offer training to livestock owners and landowners in Allan Savory's Holistic Management techniques.

Apr 4 2013: I've just looked at the managment page for the Kerr WMA http://www.tpwd.state.tx.us/huntwild/hunt/wma/find_a_wma/list/?id=12§ion=management_program
it does not specifically mention Savory or HM but there is a rotation system which involves burning, a short graze followed by 60 days rest. Quite a few other interventions seem to have been carried out. Burning isn't something I've seen mentioned by Savory.

It seems to me this is typical of lots of management. There is no one formula which can be applied world wide. Each site needs its own management regime. This might be why the academic results are mixed. Its also why a UN resolution is a bad idea. Blindly putting lots of animals on sites would do more harm than good.
• #### A comment onConversation: Is there a "breaking point" in the number of animals and land needed?

Apr 4 2013: I think that is part of his system, he uses some sort of decision management approach to find the right number of animals for the right length of time to use on the land. There is some reference to this in the comment about sigmoid in the brief Q&A at the end of the talk.

#### Favorite talks

This member doesn't have any favorite talks yet.