Témata prací (Výběr práce)Témata prací (Výběr práce)(verze: 336)
Detail práce
   Přihlásit přes CAS
Predicting Financial Market Crashes using Log-periodic Oscillation and Critical Slowing Down
Název práce v češtině: Předpovídání krachů na finančních trzích pomocí log-periodické oscilace a kritického zpomalování
Název v anglickém jazyce: Predicting Financial Market Crashes using Log-periodic Oscillation and Critical Slowing Down
Klíčová slova: Finanční trhy, Kritické body, Přechod fází, Log-periodická oscilace, Kritické zpomalování, Metody nelineární optimalizace
Klíčová slova anglicky: Financial Markets, Critical Points, Phase Transition, Log-periodic Oscillation, Critical Slowing Down, Non-linear Optimization Methods
Akademický rok vypsání: 2017/2018
Typ práce: bakalářská práce
Jazyk práce: angličtina
Ústav: Institut ekonomických studií (23-IES)
Vedoucí / školitel: prof. PhDr. Ladislav Krištoufek, Ph.D.
Řešitel: skrytý - zadáno vedoucím/školitelem
Datum přihlášení: 17.05.2018
Datum zadání: 17.05.2018
Datum a čas obhajoby: 10.06.2019 09:00
Místo konání obhajoby: Opletalova - Opletalova 26, O314, Opletalova - místn. č. 314
Datum odevzdání elektronické podoby:09.05.2019
Datum proběhlé obhajoby: 10.06.2019
Oponenti: PhDr. Oliver Polyák
Kontrola URKUND:
Seznam odborné literatury
[1] Sornette, D. (2003). Why Stock Market Crashes: Critical Events in Complex Financial Systems. Princeton. Princeton University Press.
[2] Scheffer, M. et al. Early-warning signals for critical transitions. Nature. 461, 53-59 (2009). DOI: 1038/nature08227.
[3] Scheffer, M. et al. Anticipating Critical Transitions. Science 338, 344-348 (2012). DOI: 10.1126/science.1225244
[4] Brée, D. S., Joseph, N. L. Testing for financial crashes using the Log Periodic Power Law model. International Review of Financial Analysis. 30, 289-297 (2013). DOI: 10.1016/j.irfa.2013.05.005
[5] Johansen, A. Characterization of large price variations in financial markets. Physica A. 324, 157-166 (2003).
Předběžná náplň práce v anglickém jazyce
Research question and motivation
As there is still an occurrence of financial market crashes following from deviation of security prices from their fundamental values and subsequent significant drop, a lot of questions have arisen whether and how these unpleasant events are predictable. Since the classic theory is not able to explain plummeting prices, many models from different scientific fields have been adopted.
In 2003, Sornette et al., proposed the log-periodic power law fitting the stock index prices during pre-crisis phases. Then, in 2009, Scheffer et al., studied the originally biological phenomena Critical Slowing Down describing a period of increasing variance, autocorrelation and slowing recovery of a system preceding the transition phase when the whole environment collapses and thus it can be applied to predict such downfalls.
In my bachelor’s thesis, I would like to grasp these two concepts, study the accuracy of the models and using modelling to assess their predictive power.

Since large amounts of money are invested in stock portfolios, the presence of crisis and related risk of imprecise recognition represents a true issue. Hence, my overall contribution should be to calibrate these models, apply them on real-life situations and determine their appropriatness of utilization.

I will conduct time series analysis of important world stock indices such as Standard & Poor’s 500, Dow Jones Industrial Average, Nikkei 225 or Europe Stoxx 600. In the case of Critical Slowing Down I will try to control specific characteristics, regarding Log-Periodic Power Law I will test the precision of the model proposed by Sornet or Bréé and Joseph.

1. Introduction to market crashes, their prediction and basic description our models
2. Literature review
3. Dataset – choosing data and its description
4. Methodology - Building prediction models, fitting models on the dataset
5. Results
6. Discussion and comparing the performance of both models
7. Conclusion
Univerzita Karlova | Informační systém UK