RH Logo
Hubs
About
Live
Leaderboard

My Hubs
All
ResearchHub Feeds
My Hubs
All

Trending Users

Author Profile Avatar
Colin Moser
Author Profile Avatar
Eric Cuellar
Author Profile Avatar
Emeka A. Ezewudo
Author Profile Avatar
Chukwuma Chidera
Author Profile Avatar
ayotune adebayo
Author Profile Avatar
Terry James
Author Profile Avatar
ALYSSA ENG
Author Profile Avatar
Advokat Malbuev
Author Profile Avatar
Anthony Jasmin
Author Profile Avatar
Hamza Aziz warsak

Trending Papers in Finance

Trending
Today
Trending
Today

Sign in to discover all of the research papers you care about, live as they're published.

1
Authors: G. Wang, M. Avellaneda, T. Li, A. Papanicolaou, G. Wang
Published: Mar 2021
Authors: G. Wang, M. Avellaneda, T. Li, A. Papanicolaou, G. Wang
Published: Mar 2021
We propose a new approach for trading VIX futures. We assume that the termstructure of VIX futures follows a Markov model. The trading strategy selects amulti-tenor position by maximizing the expected utility for a day-ahead horizongiven the current shape and level of the VIX futures term structure.Computationally, we model the functional dependence between the VIX futurescurves, the VIX futures positions, and the expected utility as a deep neuralnetwork with five hidden layers. Out-of-sample backtests of the VIX futurestrading strategy suggest that this approach gives rise to reasonable portfolioperformance, and to positions in which the investor can be either long or shortVIX futures contracts depending on the market environment.
Slide 1 of 1
  • Paper Preview Page 1
1
Authors: Ruiqi Liu, Vicky Henderson, Saul Jacka, Ruiqi Liu
Published: Mar 2021
Authors: Ruiqi Liu, Vicky Henderson, Saul Jacka, Ruiqi Liu
Published: Mar 2021
We study a mathematical model capturing the support/resistance line method (atechnique in technical analysis) where the underlying stock price transitionsbetween two states of nature in a path-dependent manner. For optimal stoppingproblems with respect to a general class of reward functions and dynamics,using probabilistic methods, we show that the value function is $C^1$ andsolves a general free boundary problem. Moreover, for a wide range ofutilities, we prove that the best time to buy and sell the stock is obtained bysolving free boundary problems corresponding to two linked optimal stoppingproblems. We use this to numerically compute optimal trading strategies forseveral types of dynamics and varying degrees of relative risk aversion. Wethen compare the strategies with the standard trading rule to investigate theviability of this form of technical analysis.
Slide 1 of 1
  • Paper Preview Page 1
1
Authors: John Cartlidge, Zijian Shi, Yu Chen, John Cartlidge
Published: Mar 2021
Authors: John Cartlidge, Zijian Shi, Yu Chen, John Cartlidge
Published: Mar 2021
In an order-driven financial market, the price of a financial asset isdiscovered through the interaction of orders - requests to buy or sell at aparticular price - that are posted to the public limit order book (LOB).Therefore, LOB data is extremely valuable for modelling market dynamics.However, LOB data is not freely accessible, which poses a challenge to marketparticipants and researchers wishing to exploit this information. Fortunately,trades and quotes (TAQ) data - orders arriving at the top of the LOB, andtrades executing in the market - are more readily available. In this paper, wepresent the LOB recreation model, a first attempt from a deep learningperspective to recreate the top five price levels of the LOB for small-tickstocks using only TAQ data. Volumes of orders sitting deep in the LOB arepredicted by combining outputs from: (1) a history compiler that uses a GatedRecurrent Unit (GRU) module to selectively compile prediction relevant quotehistory; (2) a market events simulator, which uses an Ordinary DifferentialEquation Recurrent Neural Network (ODE-RNN) to simulate the accumulation of netorder arrivals; and (3) a weighting scheme to adaptively combine thepredictions generated by (1) and (2). By the paradigm of transfer learning, thesource model trained on one stock can be fine-tuned to enable application toother financial assets of the same class with much lower demand on additionaldata. Comprehensive experiments conducted on two real world intraday LOBdatasets demonstrate that the proposed model can efficiently recreate the LOBwith high accuracy using only TAQ data as input.
Slide 1 of 1
  • Paper Preview Page 1
1
Authors: Rolando Martinez, Rolando Martinez
Published: Mar 2021
Authors: Rolando Martinez, Rolando Martinez
Published: Mar 2021
Benchmarks are standards that allow to identify opportunities for improvementamong comparable units. This study suggests a 2-step methodology forcalculating probabilistic benchmarks in noisy data sets: (i) double-hyperbolicundersampling filters the noise of key performance indicators (KPIs), and (ii)a relevance vector machine estimates probabilistic benchmarks with denoisedKPIs. The usefulness of the methods is illustrated with an application to adatabase of nano-finance+. The results indicate that-in the case ofnano-finance groups-a higher discrimination power is obtained with variablesthat capture the macro-economic environment of the country where a groupoperates. Also, the estimates show that groups operating in rural regions havedifferent probabilistic benchmarks, compared to groups in urban and peri-urbanareas.
Slide 1 of 1
  • Paper Preview Page 1
1
Authors: Nakagawa, Kei, et al
Published: Mar 2021
Authors: Nakagawa, Kei, et al
Published: Mar 2021
Deep hedging (Buehler et al. 2019) is a versatile framework to compute theoptimal hedging strategy of derivatives in incomplete markets. However, thisoptimal strategy is hard to train due to action dependence, that is, theappropriate hedging action at the next step depends on the current action. Toovercome this issue, we leverage the idea of a no-transaction band strategy,which is an existing technique that gives optimal hedging strategies forEuropean options and the exponential utility. We theoretically prove that thisstrategy is also optimal for a wider class of utilities and derivativesincluding exotics. Based on this result, we propose a no-transaction bandnetwork, a neural network architecture that facilitates fast training andprecise evaluation of the optimal hedging strategy. We experimentallydemonstrate that for European and lookback options, our architecture quicklyattains a better hedging strategy in comparison to a standard feed-forwardnetwork.
Slide 1 of 1
  • Paper Preview Page 1
1
Authors: Philipp Trunschke, Christian Bayer, Martin Eigel, Leon Sallandt, Philipp Trunschke
Published: Mar 2021
Authors: Philipp Trunschke, Christian Bayer, Martin Eigel, Leon Sallandt, Philipp Trunschke
Published: Mar 2021
An efficient compression technique based on hierarchical tensors for popularoption pricing methods is presented. It is shown that the "curse ofdimensionality" can be alleviated for the computation of Bermudan option priceswith the Monte Carlo least-squares approach as well as the dual martingalemethod, both using high-dimensional tensorized polynomial expansions. Thisdiscretization allows for a simple and computationally cheap evaluation ofconditional expectations. Complexity estimates are provided as well as adescription of the optimization procedures in the tensor train format.Numerical experiments illustrate the favourable accuracy of the proposedmethods. The dynamical programming method yields results comparable to recentNeural Network based methods.
Slide 1 of 1
  • Paper Preview Page 1
15
Published: Jul 2020
Published: Jul 2020
This report investigates the development of the blockchain ecosystem in Italy, against the background of the SME and entrepreneurship structure and trends in the country. The report analyses in particular the characteristics and trends of companies introducing blockchain-based services in the Italian market, opportunities and challenges to their business development, sectors and firms being targeted, and relevance for enhancing digitalisation and productivity in the SME population at large. The report also illustrates recent trends in regulation and policy, and provides policy recommendations.
1
Authors: Stefan Zeranski, Ibrahim E. Sancak
Published: Feb 2021
Authors: Stefan Zeranski, Ibrahim E. Sancak
Published: Feb 2021
Abstract The U.S. financial markets faced an unprecedented rapid decline and recovery on May 6, 2010, known as the May 6 flash crash. Roughly one trillion $ market value in less than thirty minutes vanished with the biggest one-day point decline in the history of the DJIA at the time. Since the market events took place in electronic markets, and algorithmic trading and high-frequency trading, parts of FinTech, played significant roles, we handle the May 6 flash crash from the FinTech, SupTech, and financial supervision perspectives. With the flashback method, we analyzed the reactions of market participants, media, and two financial supervisors, the SEC, and the CFTC, to the market crash. We find that the technological imbalance between financial markets or institutions and their supervisors drove the markets in uncertainty, hence in a fear and panic environment. Since the imbalance has not diminished yet, the same risks still exist. As a remedy, we introduce a new concept and model with a well-functioning SupTech system to cope with the May 6 type FinTech crises.
1
Authors: Siti Epa Hardiyanti, Lukmanul Hakim Aziz
Published: Feb 2021
Authors: Siti Epa Hardiyanti, Lukmanul Hakim Aziz
Published: Feb 2021
This study aims to investigate the impact of COVID-19 on the increase in bad credits at conventional commercial banks in Indonesia. The data used in this study are secondary data sourced from the Ministry of Health and from the Financial Services Authority (OJK), each of which consists of 50 data samples. The data analysis technique used in this study is simple regression analysis to determine the magnitude of the influence of COVID-19 on non-performing loans. The results of the data analysis show that COVID-19 has a significant effect on non-performing loans, and the COVID-19 variable can be used as an external indicator of the increase in non-performing loans for commercial banks in Indonesia. The implication of the research is that other researchers can make COVID-19 an external indicator of an emergency beyond human ability that can affect the level of non-performing loans. For banking, this study can be used as a reference when considering credit risk management policy during the COVID-19 pandemic.AcknowledgmentThe researchers are grateful to University of Sultan Ageng Tirtayasa for financial support. In addition, the authors sincerely apologize for the errors and mistakes found in this paper.
Slide 1 of 1
  • Paper Preview Page 1
1
Authors: Yannick Hoga
Published: Feb 2021
Authors: Yannick Hoga
Published: Feb 2021
Abstract Empirical evidence for multivariate stock suggests that there are changes from asymptotic independence to asymptotic dependence and vice versa. Under asymptotic independence, the probability of joint extremes vanishes, whereas under asymptotic dependence, this probability remains positive. In this paper, we propose a dynamic model for bivariate extremes that allows for smooth transitions between regimes of asymptotic independence and asymptotic dependence. In doing so, we ignore the bulk of the distribution and only model the joint tail of interest. We propose a maximum-likelihood estimator for the model parameters and demonstrate its accuracy in simulations. An empirical application to losses on the CAC 40 and DAX 30 illustrates that our model provides a detailed description of changes in the extremal dependence structure. Furthermore, we show that our model issues adequate forecasts of systemic risk, as measured by CoVaR. Finally, we find some evidence that our CoVaR forecasts outperform those of a benchmark dynamic t-copula model.returns
Load More Papers