Forums  > General  > Option pricing  
     
Page 1 of 1
Display using:  

berkalp


Total Posts: 3
Joined: Jan 2019
 
Posted: 2019-01-06 21:12
Hello!

I am studying msc financial engineering in Sweden.

I am about to select my thesis topic and there are a lot of discussion at the moment.

Some of the suggested topics are as follows:

Double stochastic option pricing, barrier option pricing, dynamic model identification for economic predictions, stochastic tensors for economic models of firms.

Me and my friends are looking at dynamic models & tensors. However we are very confused about option pricing. Is there any way that doing our thesis about options would help us finding jobs?

bullero


Total Posts: 30
Joined: Feb 2018
 
Posted: 2019-01-06 22:31
No. Learning new stuff is completely foolish.

nikol


Total Posts: 594
Joined: Jun 2005
 
Posted: 2019-01-07 00:27
Second to bullero: look into Ricci tensors, they can be useful

berkalp


Total Posts: 3
Joined: Jan 2019
 
Posted: 2019-01-09 17:30
Thanks for the replies,

@bullero, I mean, I can not see anything to learn from barrier option pricing, even if includes some deeper model. I think other topics will teach me more. But I am not sure so I need feedback.

@nikol Interesting read, I will discuss with my teacher.

I still need some feedback on the topics though.. I would be glad for replies

bullero


Total Posts: 30
Joined: Feb 2018
 
Posted: 2019-01-10 17:28
@berkalp - My first answer was meant to be sarcastic one. What I was trying to say is that it does not really matter what you do in your thesis. Do something that you think is fun and interesting. If you go to a bank they already have models that are 'cutting edge'. I remember one story of my previous colleague who joined the model team who did his PhD about pseudo random numbers (btw, I know nothing about pseudo random generators) and how this new method was supposed to be somehow 'better' compared to the existing sampling methods. It turned out that the sampling method that was developed in-house was better, in the same sense. A lot of mathematical research happens in shadows inside financial institutions and those papers are not being published. So there is a high likelihood that your research will be already outdated. So a better approach is to do something that is fun and interesting. Maybe you will learn something useful during your journey.

berkalp


Total Posts: 3
Joined: Jan 2019
 
Posted: 2019-01-10 17:37
Heh thanks a lot for the answer. I had a similar feeling about it. However I would like to land a good job & I like most of the topics, they are interesting.

I guess I ll go with dynamic model distances for economic predictions of the firms, it seems like a fun topic that will allow me to use quality check different models.

Tensors seem really hard, maybe because i am not familiar with the topic.

nikol


Total Posts: 594
Joined: Jun 2005
 
Posted: 2019-01-10 19:21
Hi, I should admit I picked up sarcastic tone from bullero and lifted it up a bit :)

I see two ways to proceed:
- purely theoretical - you read the books, pick up the (identified and unsolved) problem, crack it through hard work and deliver results
- empirical - you pick up the problem, search and download data, analyse it, discover effects, try to explain them completely yourself or through someone else theoretical work - however, in finance, high ranking analysts do both, empirical work and cook theory as well.
My personal route is always empirics.

For example, one of the first fascinating empirical facts which I saw with my eyes was an observation of pinning effect of double averaging bermudan warrant (average strike over average settlements). The price of underlying got pinned to certain level due to hedging activities and, therefore, market dried up completely. The market is full of such effects.

bullero


Total Posts: 30
Joined: Feb 2018
 
Posted: 2019-01-10 19:39
Another story from the realm of theoretical option pricing (it almost feels like I am an old dude already... and I am not even 30 yet)... While I was working for one of the bulge brackets I noticed that the pricing of baskets was pretty much driven by a technique which could be argued to be kind of...well, a "rule-of-thumb": you adjust the implied level such that your own thumb (whichever was thicker) fits between the implied and realized.

Maybe we could represent a traders thumb using tensors and then use it for pricing?
Previous Thread :: Next Thread 
Page 1 of 1