{"id":1219,"date":"2026-02-10T14:26:30","date_gmt":"2026-02-10T14:26:30","guid":{"rendered":"https:\/\/blogs.bath.ac.uk\/business-and-society\/?p=1219"},"modified":"2026-02-10T14:26:30","modified_gmt":"2026-02-10T14:26:30","slug":"phd-profile-arjun-chatterjee","status":"publish","type":"post","link":"https:\/\/blogs.bath.ac.uk\/business-and-society\/2026\/02\/10\/phd-profile-arjun-chatterjee\/","title":{"rendered":"PhD profile: Arjun Chatterjee"},"content":{"rendered":"

\"\"Arjun<\/a> has just completed a PhD focusing on how artificial intelligence can be applied in finance. He explains why passion isn\u2019t enough for a PhD, how AI can help avoid equations, and the unexpected similarities between stock markets and airflow dynamics.<\/em><\/p>\n

What drove you to choose Bath for your PhD?<\/strong><\/h3>\n

First and foremost, the thing about Bath that really caught my eye was the Doctoral College. I think a lot of people don't appreciate just how massive it is compared to other UK universities and the amount of support that you get. I remember in my orientation, I was really taken aback by exactly how much they do for doctoral students.<\/p>\n

Secondly, I just really got along with my supervisors, Professors David Newton and Dimitrios Gounopoulos \u2013 immediately, it just kind of clicked.<\/p>\n

And also, if you're going to really put your head into this for three to four years and pick a spot to study, Bath is an amazing place!<\/p>\n

 <\/p>\n

W<\/strong>hat is your research topic and how did you decide on it?<\/strong><\/h3>\n

Essentially what we're trying to do is create models for figuring out the probabilities of where asset prices might move. For example, Apple prices might be $100 today, but what are the odds they\u2019re going to be $105 tomorrow? Or $95?<\/p>\n

We can build mathematical models that dictate these underlying probabilities. However, unfortunately a lot of these models tend to be extremely difficult to work with. They don't have easy solutions, and so we almost always have to apply some sort of approximation: that's where I come in.<\/p>\n

We have been trying to use AI models to build these approximation algorithms. Neural networks are such powerful engines, which you can apply to any problem.<\/p>\n

\u00a0<\/strong><\/p>\n

What initially sparked your interest in the topic?<\/strong><\/h3>\n

My master\u2019s was in aeronautical engineering. I think that financial markets are somewhat similar to modelling airflow. It might be a little weird, but both are highly complex, non-linear systems that require extreme modelling.<\/p>\n

With finance there's an added element in the sense that if you run an experiment on airflow 100 times, it's the same result 100 times, but a stock market is going to behave differently every single time because people learn and adapt. So there's this dynamic element within it, which seemed like a really interesting problem to try to tackle.<\/p>\n

\u00a0<\/strong><\/p>\n

What are your findings so far?<\/strong><\/h3>\n

My PhD has taken the three-paper format. Within this, the first paper was a comparative study between different neural networks to identify if there are any features and certain algorithms that make them better at approximating as needed. We found there are certain models with inbuilt inductive biases, and these tend to outperform the more general models.<\/p>\n

For the second paper, we took these features and then we built our own neural network. And, amazingly, it ended up working. I think it was twice as fast as the current academic benchmark and it was a lot more accurate in certain regions of the solutions.<\/p>\n

The third paper, which I'm currently wrapping up, is trying to apply our novel algorithm to more interesting options, pricing contracts and so on.<\/p>\n

\u00a0<\/strong><\/p>\n

How could your research be applied in a real-world setting?<\/strong><\/h3>\n

I think a lot of the real-world use cases will be more for the high-frequency trading firms, hedge funds that work with low-latency trades \u2013 basically, people who are going to be trading at really high speeds.<\/p>\n

One of the main benefits from applying our method is that you can bypass having to solve an entire set of equations and get instantaneous pricing. There's a speed element as well as an ease-of-use element.<\/p>\n

We hope that it has a benefit for investors. I do think there is something in here for everybody who wants to try to trade options contracts.<\/p>\n

\u00a0<\/strong><\/p>\n

What advice would you offer to others studying or considering a PhD?<\/strong><\/h3>\n

I feel like oftentimes passion only gets you so far. I feel like it really burns for about three months in a little honeymoon phase where everybody's really excited about the PhD and wants to get on with it.<\/p>\n

It\u2019s all fine and dandy until you come across the first really difficult paper that you're trying to get through, and then everything just hits at once. You're like, \u2018Why am I here? What am I doing?\u2019 Things like passion and enjoyment can run out really quickly.<\/p>\n

Make sure you're doing the PhD for the right reasons and for yourself.<\/p>\n","protected":false},"excerpt":{"rendered":"

Arjun has just completed a PhD focusing on how artificial intelligence can be applied in finance. He explains why passion isn\u2019t enough for a PhD, how AI can help avoid equations, and the unexpected similarities between stock markets and airflow...<\/p>\n","protected":false},"author":1377,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_post_was_ever_published":false},"categories":[289,182,244,272,25],"tags":[334,277,183,382,383],"class_list":["post-1219","post","type-post","status-publish","format-standard","hentry","category-ai","category-finance","category-phd","category-phd-profile","category-technology","tag-ai","tag-artificial-intelligence","tag-finance","tag-fintech","tag-modelling"],"acf":[],"jetpack_featured_media_url":"","jetpack_shortlink":"https:\/\/wp.me\/pd4Pj1-jF","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/posts\/1219","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/users\/1377"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/comments?post=1219"}],"version-history":[{"count":0,"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/posts\/1219\/revisions"}],"wp:attachment":[{"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/media?parent=1219"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/categories?post=1219"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/business-and-society\/wp-json\/wp\/v2\/tags?post=1219"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}