Summer 2021

Google Meet link: https://meet.google.com/nvc-kbsy-ssq

Tuesday 5PM

Topic: Detection of Jump Discontinuities in Financial Time Series

Duration: 6 weeks, Starting date: Tuesday 29th June, End date: Tuesday 10th August.

The participants are supposed to read the papers in the references before starting of the summer school. It is important that they gain some level of understanding about the results presented in those. The details will be discussed during the summer project.

Reference:

  1. George J. Jiang, Estimation of Jump-Diffusion Processes Based on Indirect Inference, IFAC Computation in Economics, Finance and Engineering: Economic Systems, Cambridge, UK, 1998: 385-390.

  2. Mancini, Cecilia. Non-parametric Threshold Estimation for Models with Stochastic Diffusion Coefficient and Jumps. Scandinavian Journal of Statistics 36 (2009): 270-296.

  3. Figueroa-Lopez, Jose E.; Mancini, Cecilia. Optimum thresholding using mean and conditional mean squared error. J. Econometrics 208.1 (2019): 179-210.

  4. Das, Milan Kumar; and Goswami, Anindya. Inference of Binary Regime Models with Jump Discontinuities (2020) arxiv.org/abs/1910.10606.

Selected candidates

  1. Dhanashree Sandeep Somani

  2. Sneha Kharya

  3. Kingshuk Dutta

  4. Atharva Bhide

  5. Subham Kumar Samal

  6. Pratik Singh

  7. Namasivayam G

  8. Reetish Padhi

  9. Soumyodeep Mukhopadhyay

  10. Vinita Mukund Mulay

  11. Anirban Roy Chowdhury

  12. Kapil Chandak

  13. Vatsal Garg

  14. Priyansha Gupta

  15. Vaishnav Garg

  16. Tushar Arora

  17. Ritik Roshan Giri

  18. Amarpal Singh Basra

  19. Prakriti Barua

  20. Bihan Chatterjee

FAQ

  1. Events occurring in real life tend to have an effect on the market. So, in general, an investor tries to link most of the fluctuations in the market with some economical, political, or natural events. How can a mathematical model predict such fluctuations?
    There are two types of models of asset price dynamics. One is econometric another is mathematical.

    An econometric model relies on the history of the immediate past and predicts a range of possible values of the asset price for the next time interval with a certain level of confidence. This is basically time series analysis and is based on statistics. Such models may possibly capture usual trading behavior but cannot capture relatively rare events. Please recall that for a sufficiently liquid asset, the price changes every hour, without a need for an event to occur.

    A mathematical model on the other hand does not attempt to predict tomorrow's asset price. This rather assumes that asset prices are not predictable and uncertainties are inherent. Instead of predicting the realization of the asset price, a Mathematical model models the conditional law of the price dynamics. In other words, it attempts to answer the following question. "If the price at any future time is treated as a random variable, what is its conditional distribution given the past realization of the asset price?"

    Although the real events are not predictable, an asset price gets affected regularly by some events or the other. Although the exact occurrence of such shocks is not predictable, an empirical distribution of the shocks can be obtained and is meaningful. This distribution/law is more robust than the realizations and does not alter frequently. So, mathematical modeling of the conditional distribution of asset price     is relevant for the application.

  2. Why should one be interested in modeling the conditional distribution of asset price?
    In the market, many types of assets are traded. Some secondary risky asset prices depend on the future distribution of a primary risky asset. For finding a rational price of such secondary assets it's important to have a Mathematical model of the corresponding primary asset.

  3. I am reading a paper and I am not able to fully understand this. I need help in finding some appropriate material so that I am able to completely understand.
    Please contact me to schedule a class where you will present the concepts of the paper in simple terms to everyone and then place your question. If anybody else has also read that part and has similar or other doubts on that paper may also mention those. We will have a discussion following that.

  4. I am not familiar with the concepts of Black Scholes's option pricing theory. What should I read to get a working knowledge?
    Read sequentially

    1. https://en.wikipedia.org/wiki/Stock

    2. https://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/1997/press.html

    3. https://en.wikipedia.org/wiki/Black%E2%80%93Scholes_model

  5. I am willing to learn the subject of math finance beyond the scope of this summer school. I am seeking guidance on how to start with.
    Check this link. You may find these useful references

  6. I wish to learn more about the use of simulation for finding the theoretical price of options. How should I start?
    Read     this report made by some earlier students.

  7. I wish to learn some fundamentals of simulation techniques quickly. How should I start?
    Read     this report made by some earlier students.