Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userPredicting the Behavior of Dealers in Over-The-Counter Corporate Bond Markets
57
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Trading in Over-The-Counter (OTC) markets is facilitated by broker-dealers, in comparison to public exchanges, e.g., the New York Stock Exchange (NYSE). Dealers play an important role in stabilizing prices and providing liquidity in OTC markets. We apply machine learning methods to model and predict the trading behavior of OTC dealers for US corporate bonds. We create sequences of daily historical transaction reports for each dealer over a vocabulary of US corporate bonds. Using this history of dealer activity, we predict the future trading decisions of the dealer. We consider a range of neural network-based prediction models. We propose an extension, the Pointwise-Product ReZero (PPRZ) Transformer model, and demonstrate the improved performance of our model. We show that individual history provides the best predictive model for the most active dealers. For less active dealers, a collective model provides improved performance. Further, clustering dealers based on their similarity can improve performance. Finally, prediction accuracy varies based on the activity level of both the bond and the dealer.
rate research
Read More
The Black-Scholes Option pricing model (BSOPM) has long been in use for valuation of equity options to find the prices of stocks. In this work, using BSOPM, we have come up with a comparative analytical approach and numerical technique to find the price of call option and put option and considered these two prices as buying price and selling price of stocks of frontier markets so that we can predict the stock price (close price). Changes have been made to the model to find the parameters strike price and the time of expiration for calculating stock price of frontier markets. To verify the result obtained using modified BSOPM we have used machine learning approach using the software Rapidminer, where we have adopted different algorithms like the decision tree, ensemble learning method and neural network. It has been observed that, the prediction of close price using machine learning is very similar to the one obtained using BSOPM. Machine learning approach stands out to be a better predictor over BSOPM, because Black-Scholes-Merton equation includes risk and dividend parameter, which changes continuously. We have also numerically calculated volatility. As the prices of the stocks goes high due to overpricing, volatility increases at a tremendous rate and when volatility becomes very high market tends to fall, which can be observed and determined using our modified BSOPM. The proposed modified BSOPM has also been explained based on the analogy of Schrodinger equation (and heat equation) of quantum physics.
Electronic platform has been increasingly popular for the execution of large orders among asset managers dealing desks. Properly monitoring each individual trade by the appropriate Transaction Cost Analysis (TCA) is the first key step towards this electronic automation. One of the challenges in TCA is to build a benchmark for the expected transaction cost and to characterize the price impact of each individual trade, with given bond characteristics and market conditions. Taking the viewpoint of an investor, we provide an analytical methodology to conduct TCA in corporate bond trading. With limited liquidity of corporate bonds and patchy information available on existing trades, we manage to build a statistical model as a benchmark for effective cost and a non-parametric model for the price impact kernel. Our TCA analysis is conducted based on the TRACE Enhanced dataset and consists of four steps in two different time scales. The first step is to identify the initiator of a transaction and the riskless-principle-trades (RPTs). With the estimated initiator of each trade, the second step is to estimate the bid-ask spread and the mid-price movements. The third step is to estimate the expected average cost on a weekly basis via regularized regression analysis. The final step is to investigate each trade for the amplitude of its price impact and the price decay after the transaction for liquid corporate bonds. Here we apply a transient impact model (TIM) to estimate the price impact kernel via a non-parametric method. Our benchmark model allows for identifying and improving best practices and for enhancing objective and quantitative counter-party selections. A key discovery of our study is the need to account for a price impact asymmetry between customer-buy orders and consumer-sell orders.
This paper presents a deep learning framework based on Long Short-term Memory Network(LSTM) that predicts price movement of cryptocurrencies from trade-by-trade data. The main focus of this study is on predicting short-term price changes in a fixed time horizon from a looking back period. By carefully designing features and detailed searching for best hyper-parameters, the model is trained to achieve high performance on nearly a year of trade-by-trade data. The optimal model delivers stable high performance(over 60% accuracy) on out-of-sample test periods. In a realistic trading simulation setting, the prediction made by the model could be easily monetized. Moreover, this study shows that the LSTM model could extract universal features from trade-by-trade data, as the learned parameters well maintain their high performance on other cryptocurrency instruments that were not included in training data. This study exceeds existing researches in term of the scale and precision of data used, as well as the high prediction accuracy achieved.
In this paper, we investigate the cooling-off effect (opposite to the magnet effect) from two aspects. Firstly, from the viewpoint of dynamics, we study the existence of the cooling-off effect by following the dynamical evolution of some financial variables over a period of time before the stock price hits its limit. Secondly, from the probability perspective, we investigate, with the logit model, the existence of the cooling-off effect through analyzing the high-frequency data of all A-share common stocks traded on the Shanghai Stock Exchange and the Shenzhen Stock Exchange from 2000 to 2011 and inspecting the trading period from the opening phase prior to the moment that the stock price hits its limits. A comparison is made of the properties between up-limit hits and down-limit hits, and the possible difference will also be compared between bullish and bearish market state by dividing the whole period into three alternating bullish periods and three bearish periods. We find that the cooling-off effect emerges for both up-limit hits and down-limit hits, and the cooling-off effect of the down-limit hits is stronger than that of the up-limit hits. The difference of the cooling-off effect between bullish period and bearish period is quite modest. Moreover, we examine the sub-optimal orders effect, and infer that the professional individual investors and institutional investors play a positive role in the cooling-off effects. All these findings indicate that the price limit trading rule exerts a positive effect on maintaining the stability of the Chinese stock markets.
The distribution of the return intervals $tau$ between volatilities above a threshold $q$ for financial records has been approximated by a scaling behavior. To explore how accurate is the scaling and therefore understand the underlined non-linear mechanism, we investigate intraday datasets of 500 stocks which consist of the Standard & Poors 500 index. We show that the cumulative distribution of return intervals has systematic deviations from scaling. We support this finding by studying the m-th moment $mu_m equiv <(tau/<tau>)^m>^{1/m}$, which show a certain trend with the mean interval $<tau>$. We generate surrogate records using the Schreiber method, and find that their cumulative distributions almost collapse to a single curve and moments are almost constant for most range of $<tau>$. Those substantial differences suggest that non-linear correlations in the original volatility sequence account for the deviations from a single scaling law. We also find that the original and surrogate records exhibit slight tendencies for short and long $<tau>$, due to the discreteness and finite size effects of the records respectively. To avoid as possible those effects for testing the multiscaling behavior, we investigate the moments in the range $10<<tau>leq100$, and find the exponent $alpha$ from the power law fitting $mu_msim<tau>^alpha$ has a narrow distribution around $alpha eq0$ which depend on m for the 500 stocks. The distribution of $alpha$ for the surrogate records are very narrow and centered around $alpha=0$. This suggests that the return interval distribution exhibit multiscaling behavior due to the non-linear correlations in the original volatility.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
